Vous êtes sur la page 1sur 174

S. HRG.

1101111

THE PRESENT AND FUTURE OF


THE UNIVERSAL SERVICE FUND

HEARING
BEFORE THE

COMMITTEE ON COMMERCE,
SCIENCE, AND TRANSPORTATION
UNITED STATES SENATE
ONE HUNDRED TENTH CONGRESS
FIRST SESSION

MARCH 1, 2007

Printed for the use of the Committee on Commerce, Science, and Transportation

U.S. GOVERNMENT PRINTING OFFICE


WASHINGTON

73387 PDF

2012

For sale by the Superintendent of Documents, U.S. Government Printing Office


Internet: bookstore.gpo.gov Phone: toll free (866) 5121800; DC area (202) 5121800
Fax: (202) 5122104 Mail: Stop IDCC, Washington, DC 204020001

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00001

Fmt 5011

Sfmt 5011

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

SENATE COMMITTEE ON COMMERCE, SCIENCE, AND TRANSPORTATION


ONE HUNDRED TENTH CONGRESS
FIRST SESSION
DANIEL K. INOUYE, Hawaii, Chairman
JOHN D. ROCKEFELLER IV, West Virginia TED STEVENS, Alaska, Vice Chairman
JOHN F. KERRY, Massachusetts
JOHN MCCAIN, Arizona
TRENT LOTT, Mississippi
BYRON L. DORGAN, North Dakota
KAY BAILEY HUTCHISON, Texas
BARBARA BOXER, California
OLYMPIA J. SNOWE, Maine
BILL NELSON, Florida
GORDON H. SMITH, Oregon
MARIA CANTWELL, Washington
JOHN ENSIGN, Nevada
FRANK R. LAUTENBERG, New Jersey
JOHN E. SUNUNU, New Hampshire
MARK PRYOR, Arkansas
JIM DEMINT, South Carolina
THOMAS R. CARPER, Delaware
DAVID VITTER, Louisiana
CLAIRE MCCASKILL, Missouri
AMY KLOBUCHAR, Minnesota
JOHN THUNE, South Dakota
MARGARET L. CUMMISKY, Democratic Staff Director and Chief Counsel
LILA HARPER HELMS, Democratic Deputy Staff Director and Policy Director
MARGARET SPRING, Democratic General Counsel
CHRISTINE D. KURTH, Republican Staff Director and General Counsel
KENNETH R. NAHIGIAN, Republican Deputy Staff Director and Chief Counsel

(II)

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00002

Fmt 5904

Sfmt 5904

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

CONTENTS
Page

Hearing held on March 1, 2007 ..............................................................................


Statement of Senator Dorgan .................................................................................
Statement of Senator Inouye ..................................................................................
Prepared statement ..........................................................................................
Statement of Senator Klobuchar ............................................................................
Prepared statement ..........................................................................................
Statement of Senator Pryor ....................................................................................
Statement of Senator Rockefeller ...........................................................................
Statement of Senator Smith ...................................................................................
Statement of Senator Snowe ...................................................................................
Statement of Senator Stevens ................................................................................
Prepared statement ..........................................................................................
Statement of Senator Thune ...................................................................................

1
39
36
36
114
117
42
20
45
49
1
36
105

WITNESSES
Burke, Hon. John Downes, Board Member, Vermont Public Service Board;
Member, Federal-State Joint Board on Separations; and Member, FederalState Joint Board on Universal Service .............................................................
Prepared statement ..........................................................................................
Copps, Hon. Michael J., Commissioner, Federal Communications Committee ..
Prepared statement ..........................................................................................
Crothers, David, Executive Vice President, North Dakota Association of Telecommunications Cooperatives .............................................................................
Prepared statement ..........................................................................................
Gregg, Hon. Billy Jack, Director, Consumer Advocate Division, Public Service
Commission of West Virginia ..............................................................................
Prepared statement ..........................................................................................
Landis, Hon. Larry S., Commissioner, Indiana Utility Regulatory Commission ........................................................................................................................
Prepared statement ..........................................................................................
Massey, Richard N., Executive Vice President, Corporate Secretary, and General Counsel, Alltel Wireless ...............................................................................
Prepared statement ..........................................................................................
Simmons, W. Tom, Senior Vice President, Public Policy, Midcontinent Communications ..........................................................................................................
Prepared statement ..........................................................................................
Staihr, Ph.D., Brian K., Regulatory Economist, EmbarqTM Corporation ............
Prepared statement ..........................................................................................
Tate, Hon. Deborah Taylor, Commissioner, FCC and Chairman, Federal-State
Joint Board on Universal Service .......................................................................
Prepared statement ..........................................................................................
Tauke, Thomas J., Executive Vice President, Public Affairs, Policy and Communications, Verizon ...........................................................................................
Prepared statement ..........................................................................................

15
16
7
9
54
56
21
22
11
13
77
78
106
108
71
73
1
5
87
89

APPENDIX
Letter, dated March 8, 2007 to Hon. Daniel K. Inouye from George Strandell,
General Manager and CEO, Golden West Telecommunications ......................
Nelson, Hon. Bill, U.S. Senator from Florida, prepared statement .....................
Pollak, F.J., President and CEO, TracFone Wireless, Inc., prepared statement ......................................................................................................................
Wallace, Gary, Vice President, Corporate Relations, ATX Group, Inc., prepared statement ...................................................................................................

124
119
121
119

(III)

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00003

Fmt 5904

Sfmt 5904

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

IV
Page

Response to written questions submitted by Hon. Maria Cantwell to:


Hon. Michael J. Copps .....................................................................................
Hon. Deborah Taylor Tate ...............................................................................
Response to written questions submitted by Hon. Daniel K. Inouye to:
Hon. John Downes Burke ................................................................................
Hon. Michael J. Copps .....................................................................................
David Crothers ..................................................................................................
Hon. Billy Jack Gregg ......................................................................................
Hon. Larry S. Landis ........................................................................................
Richard N. Massey ...........................................................................................
W. Tom Simmons ..............................................................................................
Brian K. Staihr, Ph.D. .....................................................................................
Hon. Deborah Taylor Tate ...............................................................................
Thomas J. Tauke ..............................................................................................
Response to written questions submitted by Hon. Bill Nelson to:
Hon. John Downes Burke ................................................................................
Hon. Michael J. Copps .....................................................................................
David Crothers ..................................................................................................
Hon. Billy Jack Gregg ......................................................................................
Hon. Larry S. Landis ........................................................................................
Richard N. Massey ...........................................................................................
W. Tom Simmons ..............................................................................................
Brian K. Staihr, Ph.D. .....................................................................................
Hon. Deborah Taylor Tate ...............................................................................
Thomas J. Tauke ..............................................................................................

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00004

Fmt 5904

Sfmt 5904

S:\GPO\DOCS\73387.TXT

SCOM1

133
130
139
131
153
148
134
162
169
154
127
167
145
132
154
151
137
165
170
156
129
168

PsN: JACKIE

THE PRESENT AND FUTURE OF THE


UNIVERSAL SERVICE FUND
THURSDAY, MARCH 1, 2007

U.S. SENATE,
TRANSPORTATION,
Washington, DC.
The Committee met, pursuant to notice, at 10:03 a.m. in room
SR253, Russell Senate Office Building, Hon. Daniel K. Inouye,
Chairman of the Committee, presiding.
COMMITTEE

ON

COMMERCE, SCIENCE,

AND

OPENING STATEMENT OF HON. TED STEVENS,


U.S. SENATOR FROM ALASKA

Senator STEVENS [presiding]. The Chairman is stuck in traffic


and has asked me to start the hearing. I suggest we just allow the
witnesses to begin, and wait for the Chairman to make his opening
statement. Does that agree with you, Senator?
We welcome you all and look forward to your statements. We
would appreciate it if you can be as short as you desire, but all of
your statements will be printed in the record as though read.
Please. Ms. Tate?
STATEMENT OF HON. DEBORAH TAYLOR TATE,
COMMISSIONER, FCC AND CHAIRMAN, FEDERAL-STATE JOINT
BOARD ON UNIVERSAL SERVICE

Ms. TATE. Yes, good morning. Good morning, Vice Chairman Stevens and esteemed members of the Committee. Thank you all for
the honor, really, of being here today.
I know that the entire Joint Board appreciates having the opportunity to actually have a dialogue with you all. Many of you all
have been so instrumental in championing Universal Service policies for our Nation.
Last month, the FCC Commissioners testified before you and I
stated then my commitment to Universal Service, no matter where
Americans live. I re-emphasize that commitment today.
Also, this week, as you know, we celebrated the 10th anniversary
of E-Rate. And I just wanted to thank Senators Snowe and Rockefeller for their leadership and vision for generations of young Americans.
First, Id like to applaud Senator Stevens on the introduction of
the Universal Service for Americans Act, which addresses an array
of Universal Service issues, and, on the contributions side, provides
broader statutory authority for the FCC to assess both interstate
and intrastate revenues, a solution to expanding and stabilizing
the contribution base thats not available under the present Act. It
(1)

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00005

Fmt 6633

Sfmt 6633

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

2
was also, I might note, recommended unanimously by a previous
Joint Board. So, I look forward to a continued dialogue with Senators Smith and Dorgan and Pryor and other members of this committee who will be introducing legislation.
Today, obviously Im here not just as an FCC Commissioner, but
also as the Chair of the Federal-State Joint Board, a role that Im
honored to serve and take seriously, and obviously am very, very
pleased to be joined by a number of my colleagues on the Joint
Board. I want to thank them, as well as my colleagues who are not
here, for their commitment to the in-depth study of what are really
complex issues. I also appreciate our mutual desire to build a consensus to address the challenges before us. As I stated at the en
banc hearing, the good news is that I think we all truly share the
same goal; its just working on how we best reach the goals that
are set forth in the Act, given the challenges of todays ever-changing technology and, of course, the growing marketplace.
Ive seen and experienced firsthand the opportunities provided by
Universal Service in rural parts of Tennessee, probably impossible
without the Universal Service program.
In my written testimony, I provided you with an overview of the
work that weve done to date during the past year since I became
Chairman, but today I thought Id just like to focus on providing
some context for all the rest of the panel presenters that you all
will hear today regarding the growth of the Fund.
A modern and high-quality communications infrastructure is essential to ensure that all Americans, including those living in rural
areas, have access to the opportunities that broadband provides.
The Joint Board, like this committee, has renewed the debate regarding Universal Service funding for broadband in underserved
areas. However, changes in technology and increasing numbers
the numbers of carriers who are receiving Universal Service supporthave grown dramatically and place significant and increasing
pressure on the stability of the Fund, which now provides approximately $4 billion through the high-cost mechanism alone.
I brought a couple of charts that we had reviewed at our en banc.
Chart 1 shows that, since 2003, the incumbent LEC payments have
been relatively flat; and, they have actually begun to go down just
a little in recent years. On the other hand, chart 1 shows that almost all of the recent growth in the high-cost Universal Service is
largely a result of CETCs access to high-cost support.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00006

Fmt 6633

Sfmt 6633

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

Chart 2 shows that the USF payments to CETCs have been


growing at a rate of 101 percent per year since 2002. Specifically,
in 2000, CETCs received a million dollars. We expect that to be a
billion in 2006.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00007

Fmt 6633

Sfmt 6633

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

301tate4.eps

301tate3.eps

Chart 3 shows the rapid year-over-year dollar growth of CETCs.


This also highlights another problem, and that is that CETCs pres-

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00008

Fmt 6633

Sfmt 6633

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

301tate2.eps

301tate1.eps

ently, as you know, receive Universal Service support based on the


incumbent LECs embedded cost, or the per line support amount
that the incumbent LEC receives, rather than support based on
their own costs.

5
These charts show that our current high-cost mechanism is in
need of repair and revision. Discussion of this issue should not in
any way be construed as critical of the dazzling array of services
that competitors, including wireless providers, are bringing to
many parts of our country, including rural areas. However, as a
Federal official, I believe that we are called to be good stewards of
consumer dollars and the Fund.
The Chairman and others, including Verizon, CTIA, and Alltel,
and I think you will hear from them later, have proposed various
reverse auctions as a possible mechanism that could be used for
distributing high-cost support. Certainly, auctions could provide
technologically and competitively neutral means of controlling the
Funds growth and ensuring more efficient technology.
Other commenters, some of whom you will hear from today, have
discussed other tools: geospatial mapping, more targeted distribution of support, and improved data-based decisionmaking. I hope to
continue to facilitate the discussions among all of my colleagues,
while doing all we can to ensure affordable, quality services are
available to consumers, no matter where they choose to live in this
country. However, we must do so in a way that is sustainable, to
allow new generations of Americans to have access to the latest
generation of services, so that our country and our citizens can
compete in the increasingly global economy.
Thank you all, and Im pleased to answer questions after the
presentations.
[The prepared statement of Ms. Tate follows:]
PREPARED STATEMENT OF HON. DEBORAH TAYLOR TATE, COMMISSIONER, FCC
CHAIRMAN, FEDERAL-STATE JOINT BOARD ON UNIVERSAL SERVICE

AND

Good morning, Chairman Inouye, Vice Chairman Stevens, and distinguished


members of the Committee. I especially want to thank you, Chairman Inouye and
Vice Chairman Stevens, for your leadership and commitment to Universal Service.
I appreciate your invitation to participate in this hearing. It was exactly 1 month
ago that I sat at this table before you with the other members of the Federal Communications Commission (FCC or Commission). At that time, I stated my commitment to promoting the availability of quality, affordable telecommunications services
to consumersno matter where they liveacross the United States and I reemphasize that today.
I also wanted to recognize the work of this Committee on Universal Service
issues. I applaud Senator Stevens introduction of the Universal Service for Americans Act, S. 101, which addresses an array of Universal Service issues. For example,
the bill addresses Universal Service contributions by giving the Commission discretion to assess both interstate and intrastate revenuesa solution to expanding and
stabilizing the contribution base that is not available to the Commission under the
existing Act. I also look forward to working with other members of this Committee
who may be introducing legislation on universal service.
Today, I am here again not only as an FCC Commissioner, but also in my role
as Chair of the Federal-State Joint Board on Universal Service (Joint Board), a role
that I am honored to serve and greatly respect. I am pleased that I am joined by
some of my Joint Board colleaguesfellow FCC colleague Commissioner Mike
Copps, Commissioner Larry Landis of Indiana, Commissioner John Burke of
Vermont, and Director Billy Jack Gregg of the Consumer Advocate Division of West
Virginia. All of the Joint Board membersthose here today, as well as FCC Chairman Martin, Joint Board State Chair Ray Baum of Oregon, and Commissioner
Edgar from Floridadeserve praise for their commitment to the in-depth study of
these complex issues in addition to their full time jobs as government officials. I also
appreciate our mutual desire to build consensus to address the challenges before us.
Congress required the FCC to institute a Joint Board to recommend changes to
any of [the FCCs] regulations in order to implement sections 214(e) and [254] of
the Act. Accordingly, I welcome the opportunity to hear directly from you regarding

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00009

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

6
Universal Service issues facing the FCC, the industries we impact and most importantly, as section 254 of the Act states, consumers in all regions of the Nation.
Like many of you, I have seen and experienced firsthand the opportunities provided
by Universal Service in very rural areas. I remember the day the telephone wire
was rolled up a gravel road to my grandmothers house in rural Tennesseelikely
an impossibility without a Universal Service program. At the same time, in my roles
at the FCC and on the Joint Board, I have a responsibility to preserve and advance
the Universal Service Fund to best serve the public interest.
Since becoming Chair of the Federal-State Joint Board on Universal Service, the
Joint Board has continued its work to review the Universal Service policies and respond to the FCCs referrals. I have been committed to keeping our work on a timetable paced to fulfill our statutory role in a thoughtful and deliberative manner, including holding meetings and conference calls, issuing notices, and reviewing comments. The Joint Board staff held a retreat for 3 days in June 2006 to review outstanding and new proposals, and the Joint Board met in August 2006 during the
NARUC meeting in San Francisco. Because there were several newer members of
the Joint Board, including myself, in September, we hosted a 2-day meeting at the
FCC focusing on training. We heard from USAC, NECA and FCC Bureau experts
about the mechanics of the Universal Service programs. The state members of the
Joint Board and staff met again in November 2006 during the NARUC meeting in
Miami. The full Joint Board held its recent en banc hearing less than 2 weeks ago
here in Washington, D.C. We were pleased that members of your staffs attended as
well.
We continue to evaluate the issues expressly delegated by the FCC to the Joint
Board for consideration, including what many call the rural review proceeding and
the basis of support elements of the competitive ETC review. As a part of its analysis, the Joint Board is looking at ways to improve the distribution of high-cost Universal Service support. Accordingly, we continue to evaluate draft proposals, hear
from experts, and explore solutions that will help sustain the benefits of the Universal Service program for years to come. As Chairman of the Joint Board, I hope
to encourage discussion among my colleagues and facilitate consensus that will ensure that American consumers throughout the Nation continue to have access to an
evolving level of innovative services.
Although the Joint Board has been considering several options, last summer, the
Joint Board sought public comment on the use of reverse auctions as a tool to improve the distribution of high-cost support. On August 11, the Joint Board issued
a Public Notice and sought comment on primary questions, such as the overall appropriateness and legality of implementing reverse auctions, as well as questions
about the mechanics of any reverse auctions, such as Federal and state jurisdictional roles, quality of service obligations, and the unique questions regarding the
treatment of incumbent local exchange carriers (LECs). The Joint Board received
numerous comments and reply comments last fall, and also received additional submissions in the record. Further, as a part of last weeks en banc hearing, the Joint
Board heard experts, including witnesses from the National Telecommunications Cooperative Association, Verizon, and CTIAThe Wireless Association discussing
specific proposals, benefits, and concerns regarding the use of reverse auctions. We
also heard from experts on geo-spatial mapping and more targeted approaches to
the distribution of support that would modify our current programs, including witnesses from the Polis Center in Indianapolis, CostQuest Associates, and Embarq TM.
I am encouraged that you plan to hear from some of these same groups later today.
I think it is important to understand how technological change in the industry
is impacting the policy discussion. The communications marketplace continues to
evolve daily, as convergence shakes the foundations of the old order for industry,
for government, and for consumers alike. While this convergence creates real benefits for consumers through the introduction of exciting new services and increased
competition among multiple service providers, it also challenges us to adapt our regulations to keep pace with these technological changes.
The Joint Board continues to carefully evaluate the balance of issues at the intersection where the policies of Universal Service and competition meet. It is critical
that we not lose sight of the Universal Service goals, as we look forward to ensuring
that an evolving level of communications services are rolled out to all areas of the
country.
As we heard at the en banc, the area of greatest growth in the high-cost program
relates to the increasing entry of competitive ETCs into rural areas. The fact is that
overall support funding for incumbent LECs has been flat or decreasing in recent
years. On the other hand, we have witnessed rapid growth in the funding of
CETCssometimes funding a second, third or more entrants in what have been determined to be high-cost markets. According to FCC and USAC data, competitive

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00010

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

7
ETC funding has grown from $1 million in 2000 to $1 billion in 2006. If this continues at the present rate, CETC funding could double by funding year 2008.
This growth is not only due to multiple providers receiving high-cost support, but
also because CETCs receive Universal Service support based on the incumbent
LECs embedded costs or the per line support amount that the incumbent LEC receives. But as we heard at the en banc, as competitors enter areas supported by
Universal Service high-cost funding, their actual costs are likely to be very different,
often lower, than the incumbent telephone carriers costs on a per line basis.
Discussion of this issue should not be construed as critical of the dazzling array
of services that competitors, including wireless providers, are bringing to the rural
areas of our country. Indeed, wireless services have added a new dimension to
connectivitymobilitythat is very important to many consumers. It is no wonder
that wireless telephone connections now far out strip the number of wireline connectionsby over 25 percent, according to the FCCs most recent figures. I have mentioned the issue to you in detail because the fact is that the growth of Universal
Service high-cost support is easily identified, and is expected to continue to grow
rapidly.
As we look ahead to the long-term goals of the Universal Service program, we
must balance the goal of encouraging competitive entry with the other challenges,
such as the further deployment of advanced services. For instance, Alltel recently
filed a novel proposal to allocate funding for broadband in unserved areas through
competitive bidding. It is essential that as the converging communications landscape
changes, we recognize how technological changes are putting strains on the mechanics of our contribution and distribution systems which must be addressed by policies
that avoid subjecting the program to unsustainable growth. Like you, as a Federal
official, we are stewards of these consumer dollars. While doing all we can to ensure
that affordable, quality services are available to consumers all across the country,
we must do so in a way that is sustainable to allow new generations of Americans
to have access to the latest generation of services so that our country is able to compete in the increasingly global economy.
Again, I appreciate your invitation to be here with you today. I look forward to
hearing from you today and in the future, and I will be pleased to answer any questions.

Senator STEVENS. Well, thank you, Commissioner Tate.


Our next witness is Commissioner Michael Copps, of the FCC.
STATEMENT OF HON. MICHAEL J. COPPS, COMMISSIONER,
FEDERAL COMMUNICATIONS COMMITTEE

Mr. COPPS. Thank you, Mr. Vice Chairman, members of the Committee.
Im pleased to visit with you again today to focus on the challenge of how to bring advanced telecommunication services to all
of our citizens and to ensure that our Universal Service system,
which has accomplished so much, can make this happen in a sustainable way. Each and every citizen of this great country should
have access to the wonders of communications, whether they live
in rural areas, on tribal lands, or in our inner cities, whether they
have limited incomes or disabilities, whether they are schoolchildren or rural healthcare providers.
If were going to ensure that no community and no citizen is left
behind by a lack of access to basic or advanced telecommunications
in this new digital age, we need to make some changes. We must,
first of all, include these new opportunity-creating technologies as
part of our Universal Service program. In plainer English, it is
time to bring broadband into the Universal Service system. Then
we must fine-tune the Fund. We must broaden the USF contribution base. We must make sure funds are distributed with maximum equity among consumers, areas, and technologies. We must
fund what is necessary to achieve our goal, and no more. And we

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00011

Fmt 6633

Sfmt 6601

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

8
must appreciate that the economies of nonrural, rural, and truly
remote service areas can be very different.
If were going to broaden the purposes of Universal Service, we
need to make that commitment up front. USF surely cries out for
changes in many aspects, but isnt it better to get the mission clear
before we do a lot of tinkering around the edges? Im not suggesting delaying the fine-tuning. Im just suggesting the urgency of
stating the message and the mission. It strikes me that first you
have an objective, then you have a program. A USF commitment
to broadband strikes me as a pressing national need. Broadband is
the great network and infrastructure challenge of our time, just
like canals and railroads and highways were in an earlier era. Our
future will, in significant measure, be decided by how well we build
our broadband connectivity in the digital age.
So, first we need to look at what role Universal Service should
play in meeting this great infrastructure challenge. I recognize that
the process of incorporating broadband will involve complex and
difficult choices about what mix of technologies, like wireless and
copper-based, and fiber, to support, and how to support them, and
over what time frame. And I dont have a silver-bullet answer, but
Im not sure anyone else does either. I do know that we need to
confront these questions in a forthright and honest fashion. We
need to resolve them through a process that involves all the stakeholders in this important issue. That surely includes the state authorities, like the experts sitting beside me here today, who are
such a fountain of creative and insightful ideas on the subject. And
I hope the FCC will play a more proactive role in the effort, not
least by gathering the hard data that is absolutely essential for
sound policymaking, doing the analyses, and teeing up options for
you and Congress to look at. You should push us to do more, much
more, in this regard.
We also need direction on whether Universal Service is going to
be the vehicle or a vehicle in a comprehensive national
broadband strategy, because such a strategy might involve additional components, like matching grants or tax incentives. But this
much I know: we simply cannot throw up our hands and say that
there shouldnt be any Federal Universal Service support for
broadband. Unfortunately, in too may ways, thats exactly what our
approach to Universal Service does today.
In truth, I believe that Congress already gave the FCC and the
states a statutory mandate to bring access to advanced telecommunications to each and every citizen of this country. Im not
sure, however, that all of my colleagues on the Commission agree
that we have the authority to include broadband in universal service, or even on whether doing so is the way to go; hence, the apparent need for Congressional guidance.
I realize much of our discussion today may be considerably more
nitty-gritty and mechanics-oriented than what Ive just said. We
have a duty to deal with the nuts and bolts of managing the program we have today. So, permit me, quickly, to propose three
things that I think could be done immediately to put Universal
Service on a more solid footing so that it can be better deployed to
help shape our future.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00012

Fmt 6633

Sfmt 6601

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

9
First, with boundaries between local and long distance eroding
and the skyrocketing success of any-distance calling plans, assessing Universal Service contributions only on interstate services is
anachronistic. While it will require a legislative fix, I believe that
assessing both intrastate and interstate revenues is a good idea. It
would significantly lower the contribution factor and would expand
the base of future funding for broadband buildout, if that is the
road you choose.
Second, it is as clear as clear can be that the costs of investing
and maintaining wireless and wireline infrastructure are inherently different. I believe that wireless can and should be part of
Universal Service, but the time has come to put an end to the irrational and costly system of supporting wireless carriers based on
the cost of wireline incumbents. The identical support rule is the
subject of a five year old Joint Board referral. I believe it is time
for the Board to make a recommendation to the full Commission.
Third, I believe that the Universal Service system cannot thrive
without regular review and care. The high-cost fund, like many
other good programs, can only benefit from additional oversight
and auditing to ensure that a few bad actors dont jeopardize the
strength of this great enabling program.
The Joint Board and the FCC are discussing how best to shore
up the Fund. Board Chairman Tate and our state colleagues here
this morning are working hard to develop recommendations for the
Commission. Our state colleagues on this mornings panel are
among the Nations leading experts on Universal Service. They
have put creative ideas before the Joint Board and the Commission. And Commissioner Tate and I may well be asked to vote on
these ideas in the months ahead.
Last week, the Joint Board held a valuable en banc hearing addressing some of the issues we will be discussing this morning. I
would like to at least see some recommendations come forth in the
next few months. And, by way of suggestion, I would hope that future referrals from the Commission would contain some time limitations for Board action. The USF can do great things for America.
It can help ensure that often unserved areas of our country are
connected to vital education, public health, public safety, employment, and business opportunities, but we do not have the luxury
of time to get this right, because the rest of the world has no intention of waiting for us.
So, thank you for holding this hearing. I look forward to our conversation today to see how we can best maintain a robust and effective and forward-looking Universal Service system that remains
true to its essential mission and true to the mission of our country.
[The prepared statement of Mr. Copps follows:]
PREPARED STATEMENT OF HON. MICHAEL J. COPPS, COMMISSIONER,
FEDERAL COMMUNICATIONS COMMISSION
Mr. Chairman, Mr. Vice Chairman, members of the Committee, I am pleased to
visit with you again today to focus on one of the most important challenges confronting this Committee, our Commission and the country. This is the challenge to
bring advanced telecommunications to all our citizens and to ensure that our Universal Service system, which has accomplished so much, can make this happen in
a sustainable way. Since I went to the FCC nearly 6 years ago, my overriding objective has been to help bring the best, most accessible and cost-effective communica-

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00013

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

10
tions system in the world to all our peopleand I always underline the all. Each
and every citizen of this great country should have access to the wonders of communicationswhether they live in rural areas, on tribal lands, or in our inner cities;
whether they have limited incomes or disabilities; whether they are schoolchildren
or rural healthcare providers.
If we are going to ensure that no community, no citizen, is left behind by lack
of access to basic or advanced telecommunications in this new digital age, we need
to think anew, adjust our policies and craft the proper incentives. We must include
these new opportunity-creating technologies as part of our Universal Service program. In plainer English, it is time to bring broadband into the Universal Service
system. We must also update and broaden the USF contribution base. We must
make sure funds are distributed with maximum equity among consumers, areas and
technologies. And we must recognize that the economics of non-rural, rural and
truly remote service areas are fundamentally different.
Permit me to begin by emphasizing the importance of an USF commitment to
broadband because this is, far and away, the most meaningful step we can take to
create opportunity for our citizens, to ensure community development in every area
of our country and to keep our Nation competitive in the global economy. Broadband
is the great network and infrastructure challenge of our time. If you double back
through the years of this Nations history, you will find that just about every formative era has had its own major infrastructure challenge. Go back to the very beginning as settlers pushed into the frontier and populated new lands. Their infrastructure challenge was to develop ways to deliver their produce and products to increasingly far-away markets. So they found ways to build roads and turnpikes and canals
and ports to meet that challenge. Later, as we industrialized, the need was to lay
a railway grid, first across regions and then across the country, climaxed by the
great saga of the Transcontinental railroads as we became a continental power following the Civil War. Closer to our own era, in the Eisenhower years as suburbs
grew and our demography changed, came the Interstate Highway System binding
the country more closely together. We saw it in communications, too, in extending
telephone service to rural America with the Rural Electrification amendments under
Harry Truman and with the Universal Service Fund that we are gathered here to
discuss this morning. In all of these infrastructure build-outs, there was a critical
role for government, business and local community organizations to work together
toward a great national objective. This is really the American Story. Its how we
built our Nation and how we grew. It is, I believe, the only way we will continue
to grow it.
From where I sit, broadband networks are the canals and railroads and highways
of the digital age. Our future will be in significant measure decided by how we master, or fail to master, advanced communications networks and how quickly and how
well we build out broadband connectivity.
So first we need to look at what part Universal Service should play in meeting
this great infrastructure challenge. I recognize that the process of incorporating
broadband into Universal Service will involve many complex and difficult choices
about what mix of technologieslike wireless, copper-based, and fiberto support,
how to support them, and on what time frame. I certainly dont have a silver bullet
answer here today, and I am not sure that anyone else does either. But I do know
that we need to confront these questions in a forthright and honest fashion, and we
need to resolve them through a process that involves all the stakeholders in this
important issue. That surely includes the state authorities, like those sitting beside
me here today, who are a fountain of creative and insightful ideas on this subject.
I hope the FCC will play a more proactive role in this effortnot least by gathering
the hard data that is absolutely essential to sound policymaking, doing the analysis
and teeing up options for Congress to consider. We also need to make sure that decisions about Universal Service are part of a complete national broadband strategy,
which might involve additional components such as matching grants and tax incentives. More than anything else, I know that we simply cant throw up our hands
and say that there shouldnt be any Federal Universal Service support for
broadband. Yet in too many ways that is exactly what our approach to Universal
Service does today.
In truth, I believe that Congress already gave the FCC and the states the statutory mandate to advance the cause of bringing access to advanced telecommunications to each and every citizen of our country. Im not sure, however, that all my
colleagues on the Commission agree that we have the authority to be more proactive
in encouraging broadband deployment and penetration, and this is why I am hopeful that Congress will choose to make this clear for all of us to understand.
Earlier this year I was fortunate enough to meet a small business owner who was
able to work out of his home on a rural hilltop on the Big Island of Hawaii after

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00014

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

11
broadband service was installedrather than trekking each day to the nearest town
miles away to get online. And not too long ago I visited an Inuit village in Alaska,
totally unreachable by road, where a sick child with an ear infection could be examined by a doctor hundreds of miles away. In another Alaskan village, students had
used their broadband connection to speak in real time with the crew of the International Space Station. Like a string wrapped around a finger, stories like these remind us that lives and livelihoods and our very health are hugely influenced by the
communications infrastructure available to us.
As we work on implementing these lofty concepts, we must also of course deal
with nitty-gritty of administering the program we have today. Permit me propose
three things that I believe could be done immediately to put Universal Service on
a more solid footing so that it can be better deployed to shape our future. First, with
boundaries between local and long distance eroding, and the skyrocketing success
of any-distance calling plans, assessing Universal Service contributions only on
interstate services is anachronistic. While it will require a legislative fix, I believe
that assessing both intrastate and interstate revenues is a good idea. Second, it is
as clear as clear can be that the costs of investing and maintaining wireless and
wireline infrastructure are inherently different. I believe that wireless can and
should be a part of Universal Service, but the time has come to put an end to the
irrational and costly system of supporting wireless carriers based on the cost of
wireline incumbents. The identical support rule is the subject of a 5-year old Joint
Board referral; I believe it is high time for the Board to make a recommendation
to the full Commission so we can take corrective action. Finally, I believe that the
Universal Service system cannot thrive without regular review and care. The highcost fund, like many other good programs, can only benefit from additional oversight
and auditing to ensure that a few bad actors do not jeopardize the strength of this
great enabling program.
The Joint Board and the FCC are in the midst of a serious debate on how to best
shore up the Universal Service Fund and how it can meet the changing needs of
the country as we head into the 21st century. Board Chairman Tate and our state
colleagues here this morning are hard at work developing recommendations for the
Commission. Our state colleagues on this panel are among the Nations leading experts on Universal Service. They have put creative ideas before the Joint Board and
Commissioner Tate and I may well be asked to vote on these ideas in the months
ahead. Last week, the Joint Board held a valuable en banc hearing addressing some
of the issues we will be discussing today. I continue to urge my colleagues that we
act quickly and deliberately to address the rising demands on Universal Service. All
of us want this system to work. None of us wants our country, or anyone in it, to
miss the opportunities of the digital age. None of us wants to see any kind of digital
gap anywhere in America. But, truth is, if we dont get our policies right, we could
experience a 21st Century Digital Gap, in spite of the wonder of all these new technologies, greater than the one we experienced with plain old telephone service in
the last century. The USF can do great things for America. It can help ensure that
often unserved areas of our country are connected to vital education, public health,
public safety, employment, and business opportunities. But we dont have the luxury
of time to get this right because the rest of the world isnt planning on waiting for
us.
I look forward to our conversation today to see how we maintain a robust, effective, and forward-looking Universal Service System that remains true to its essential mission and true to the mission of our country.
Thank you for your attention and for holding this hearing today.

The CHAIRMAN [presiding]. I thank you very much, Commissioner Copps.


Our next witness is a member of the Indiana Utility Regulatory
Commission, the Honorable Larry S. Landis.
Commissioner Landis?
STATEMENT OF HON. LARRY S. LANDIS, COMMISSIONER,
INDIANA UTILITY REGULATORY COMMISSION

Mr. LANDIS. Thank you, Mr. Chairman, Mr. Vice Chairman,


members of the Committee.
Senator Inouye, I had the privilege of visiting your state this
past year to witness the installation of my friend Chad Miles, CEO

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00015

Fmt 6633

Sfmt 6601

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

12
of a small, but highly innovative, rural company, as President of
OPASTCO.
Senator Stevens, I bring you special greetings from the city of
your birth, Indianapolis. I also had the privilege of visiting your
state last year, and had a chance to experience firsthand the
unique challenges and opportunities which both traditional and advanced communications hold out for the people of Alaska.
I thank you for the opportunity to address the critical issues relating to Universal Service from the perspective of the Joint Board
and the perspective of state regulators.
I want to underscore that I do not necessarily represent the
views of all state regulators, which, like those of this body, sometimes diverge.
Given time constraints, Id like to start by referencing the March
2, 2006, testimony of my colleague, North Dakota Public Service
Commission Chairman Tony Clark, almost exactly a year ago
today, on behalf of NARUC. His observations are still relevant.
Today, we speak more about the distribution side of Universal
Service. However, I do want to acknowledge the bipartisan effort
which went into framing a solution to funding of Universal Service,
which was incorporated into the proposed Communications Act of
2006 last year, and, as Commissioner Tate has already mentioned,
is incorporated into a freestanding bill and other legislation again
this year. The latitude which you incorporated into that plan, from
a funding perspective, was useful, commendable, and, I believe, enjoys broad support from state regulators.
I want to limit my remarks about the important issue of the significant growth in the size of the high-cost fund. I share the opinion of my colleagues on the need for a cap on expenditures to give
us breathing room to address the issues in a more comprehensive
manner. Its critically important to the sustainability of the program and to its continued place on the public policy agenda. Chairman Martin has spoken to the issue forcefully. And my Federal and
State colleagues have addressed, and will address, that issue here
today.
In considering reform, we would do well to take a page from the
Hippocratic Oath and first resolve to do no harm. Given the size,
scope, and complexity of the current mechanisms, this is a considerable challenge. For example, high-cost loop support for rural companies is currently determined based on legacy investments; or, put
another way, embedded costs. We need only look to Detroit to see
the problems which legacy decisions can present for companies
looking to move into the 21st century and to compete with companies which are not saddled with those decisions; decisions which
seemed appropriate at the time, but now may create a challenge,
a significant burden of competitive disadvantage. So, we need to
encourage companies to look to the future rather than to a legacy
past. But if we decide to sever those links to a legacy past, we also
have a responsibility to offer a reasonable migration path to those
companies which have based their business plans on that model,
which we now may consider less relevant.
Another example may be found in the challenge presented by the
growth in the number of competitive ETCs, primarily wireless companies. The FCCs guidelines on CETC designations, adopted in

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00016

Fmt 6633

Sfmt 6601

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

13
2005 in response to a recommendation of this Board, define criteria
and urge states to apply a public-interest standard. In Indiana,
weve taken this challenge seriously. Other states have chosen a
more permissive approach or, as is the case in North Dakota, were
restricted in their ability to review ETC applications by a court decision. Those 2005 guidelines should be made mandatory, and, as
states, we need to assume our share of responsibility.
At the same time, there are many rural areas where multiple
wireless providers are active, where there is already competition.
We need to make sure that we dont inadvertently advantage one
company over the others which entered that market based on a
competitive unsubsidized model.
Lurking just around the corner is the question of broadband
buildout. The problem is that there is relatively little granular data
which would tell us what form and how much should be devoted
to buildout in those high-cost and very high-cost areas. Commissioner Copps has spoken to the need for better, more robust data,
and I share and echo his concern.
I believe that states have an important and potentially growing
partner role with the FCC as joint stewards in implementing your
vision and seeing to it that Universal Service funds are appropriately disbursed, the legitimate needs are met, but that accountability and performance are audited and demanded.
Again, I thank you for the privilege of sharing our thoughts with
you this morning. I look forward to any questions you may have.
[The prepared statement of Mr. Landis follows:]
PREPARED STATEMENT OF HON. LARRY S. LANDIS, COMMISSIONER,
INDIANA UTILITY REGULATORY COMMISSION
Good morning, Mr. Chairman, Co-Chairman Stevens, and members of the Committee. I am Larry Landis, and I am a member of the Indiana Utility Regulatory
Commission. I serve on the Telecommunications Committee of the National Association of Utility Regulatory Commissioners, NARUC, and was Vice Chair of NARUCs
Intercarrier Compensation Task Force. I am also a member of the Federal-State
Joint Conference on Advanced Telecommunications Services; and most pertinent to
todays hearing, a member of the Federal-State Joint Board on Universal Service.
Thank you for the opportunity to address the critical issues relating to Universal
Service from the perspective of state regulators. I want to underscore that I do not
necessarily represent their views, which like those of this body, sometimes diverge.
Given todays time constraints, I would start by referring you back to the March
2, 2006 testimony of my colleague Tony Clark, Chairman of the NARUC Telecommunications Committee and of the North Dakota Public Service Commission, before this Committee almost exactly a year ago today. Commissioner Clarks observations then are still relevant today.
Commissioner Clark characterized Universal Service as being at a crossroads.
Among the questions he posed:
Should broadband infrastructure and services be explicitly funded?
What is the optimal size of the Fund and does it need to be capped?
Should it fund competition in high-cost markets?
How many networks should be funded in high-cost markets?
On what cost basis should carriers be reimbursed?
How many access lines per customeror householdshould be funded?
Is it intended for networks or for individuals?
Should contributions be pegged to network usage, use of numbers, connections
or some other methodology?
Should Universal Service continue to be a shared Federal-State responsibility,
or is there some other configuration which makes sense?

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00017

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

14
Today we speak more about the distribution side of Universal Service. However,
I do want to acknowledge the bipartisan effort which went into framing a solution
to funding of Universal Service which was incorporated into the proposed Communications Act of 2006 last year, and which I understand is incorporated into a freestanding bill again this year. The latitude which you incorporated into that plan
from a funding perspective was useful, commendable, and I believe enjoys broad
support from state regulators.
I will limit my remarks about the important issue of the significant growth in the
size of the high-cost funds. I share the opinion of my colleagues on the need for a
cap on expenditures to give us breathing room to address the issues in a more comprehensive way. It is critically important to the sustainability of the program and
to its continued place on the public policy agenda. Chairman Martin has spoken to
this issue forcefully and my Federal and state colleagues have addressed and will
address that issue here today.
In considering reform, we would do well to take a page from the Hippocratic oath
and first resolve to do no harm. Given the size, scope and complexity of the current
mechanisms, that is a considerable challenge.
For example, high-cost loop support for rural companies is currently determined
based on legacy investments, or put another way, embedded costs. We need only
look to Detroit to see the problems which legacy decisions can present for companies
which are looking to move into the 21st century and to compete with companies
which are not saddled with those decisions. Those decisions seemed appropriate at
the time, but now create a significant burden of competitive disadvantage.
Increasingly, facilities-based competition is coming to many rural local exchange
companies. It is coming not only in the form of mobile wireless, but also VoIP delivered by cable modem, fixed wireless and broadband over power lines. But that competition is taking root primarily in the villages, communities, towns and small cities
in those rural service areas. Often it doesnt reach out to the truly rural areas
served by rural LECs.
We need to encourage incumbentsindeed, all providersto look to the future
rather than to a legacy past. But if we decide to sever those links to a legacy past
for the RLECs, we also have a responsibility to migrate those companies which have
based their business plans on a model which we may now consider less relevant.
And we need to focus support in those areas where the costs are higher by an order
of magnitude, and which in many cases are not contestable.
Another example may be found in the challenge presented by the growth in the
number of competitive ETCs, primarily wireless companies. Some will assert that
this growth is symptomatic of the problems of Universal Service. Others will argue
that this is a reflection of the dynamic growth of the wireless sector. Regardless,
the wireless sector has been the primary contributor to growth in the high-cost
funds.
Under Section 214(e) of TA 96, State Commissions are delegated to help administer the Universal Service Fund by designating those companies which are eligible
to receive support (Eligible Telecommunications Carriers, or ETCs) in each state.
The FCCs guidelines on ETC designations and certifications, adopted in 2005 in response to a recommendation of this Board, define criteria and urge states to apply
a public interest standard.
In Indiana, we have taken this charge seriously by fully adopting these guidelines
and applying them to each new ETC applicant and each ETC who seeks annual certification for Federal Universal Service Funds. Other states have chosen a more permissive approach oras is the case in North Dakotawere restricted in their ability to review ETC applications by a court decision. Those 2005 FCC guidelines
should be made mandatory, and as states we must shoulder our share of responsibility.
At the same time, there many rural areas where multiple wireless providers are
active. Some companies have entered some rural markets based at least in part on
the assumption that they could receive Universal Service support. Other companies
have entered rural markets based on a competitive, unsubsidized model. One proposal would hold reverse auctions in those areas where there are multiple wireless
ETCs, with wireless ETC funding distributed on a winner-take-all basis. Where
there is already competition, we need to make sure we dont inadvertently advantage one company over its competitors, which entered that market based on their
assumption that it was contestable. Put another way, we need to make sure we are
not inadvertently making it more difficult to compete, thereby perhaps reducing
competition while reforming Universal Service subsidies.
Lurking just around the corner is the question of rural broadband buildout. The
problem is that there is relatively little granular data which would tell us which
of several solutions would be most cost-efficient in addressing the needs of the

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00018

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

15
unserved in any given geographical area. Once we know that, we are in a far better
position to determine what form and how much should be devoted to buildout in
those high-cost and very high-cost areas.
Where will tax abatements be sufficient incentive to encourage buildout? Where
are costs so high that only a straight subsidy will work? In the latter cases, where
the market isnt there, who will choose which technology is selected, and how large
should the subsidy be? Commissioner Copps has spoken to the need for better, more
robust data, and I share his concern.
I believe the states have an important and potentially growing partner role with
the FCC as joint stewards in implementing your vision, and in seeing to it that Universal Service funds are appropriately deployed, that legitimate needs are met, but
that accountability and performance are audited and demanded.

The CHAIRMAN. I thank you very much, Commissioner Landis.


And now, may I call upon a member of the Vermont Public Service Board, the Honorable John D. Burke.
STATEMENT OF HON. JOHN DOWNES BURKE, BOARD MEMBER,
VERMONT PUBLIC SERVICE BOARD; MEMBER, FEDERALSTATE JOINT BOARD ON SEPARATIONS; AND MEMBER,
FEDERAL-STATE JOINT BOARD ON UNIVERSAL SERVICE

Mr. BURKE. Thank you very much, Mr. Chairman, Mr. Vice
Chairman, members of the Committee. Thank you for giving us a
chance to express to you our views, our beliefs, our desires, and our
hopes for the Fund, going forward.
Theres no doubt that the stated purpose of the Universal Service
Fund, as stated in section 254(b)(3), was to provide comparable
services at comparable rates to the higher-cost areas of America. I
certainly hope that as we go forward we try to remember that is
the primary goal that all of us have in mind.
The task becomes daunting. And one of the main reasons its become daunting is in order to try to further that goal, the Fund has
become substantially larger and more inclusive than it might have
been originally anticipated to be. Theres a slide and a graph, that
we also used at the en banc meeting, that gives you an idea of
where the pressure on the Fund really presently comes from.
Commissioner Tate mentioned the fact that there is a substantial
growth in the CETC side of the Fund, the competitive side of the
Fund. If you look at that growth, you can see that its risen to an
amount of almost a billion dollars. The idea of the growth in CETC
does not end there. There are estimates and reasonable projections,
part of which were part of our en banc presentation, that would indicate that by 2009 this particular portion of the Fund may have
risen to as much as $2.5 billion, making the Fund no longer a $4
billion project, but a $6.5 billion project. Its the challenge of all of
usyou, as legislators, us, as members of the Joint Board, recommending to the FCC, and, of course, the FCCto do what we can
to try to take pressure off that fund to the point that we are able
to serve the people that need to be served, as defined in 254(b)(3),
without getting to the point of the Fund getting so large that it implodes on itself.
The graphits not meant to be exact; obviously, estimates are
estimatesbut you can see that the pattern of growth is actually
just following the trend thats existed for the past 4 years, up until
today, and going forward for the next 2.
I also would like to take this opportunity to indicate that I have
proposed a cap on the CETC side of the Fund, presumably with in-

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00019

Fmt 6633

Sfmt 6601

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

16
flation adders to allow for the Joint Board and the Congress to consider alternatives that would take the pressure off the Fund and
still allow it to handle its intended purposes under section
254(b)(3). I would hope that you would look at these alternatives
and work with us in formulating methods of going forward to both
control the Fund and allow it to succeed.
As we consider these alternatives, though, we should understand
that there are other ideas and other concerns that exist with regard to supported services under the Fund. I agree with Commissioner Copps that broadband, especially in rural America, those
areas that are harder to serve, is truly a crying need. In my own
State, although it would appear that were about 70-plus percent
served by broadband, truthfully we only have one really major metropolitan area. If we remove Burlington from the mix, we really are
less than 50 percent. For people today in rural areas, trying to do
business on the web, or trying to make a living on the web, trying
to keep informed on the web, broadband is truly a necessity, and
I really believe that, as a supported service, it could move forward.
I applaud Senator Stevens for his idea and his concerns expressed with regard to broadband in his bill. I think that we have
to move forward in broadband in ways that are creative, maybe
with matching grants or in ways, with the states, that would allow
the states to help target those areas in broadband most in need,
and also help limit the size and the pressure on the Fund by having the states participate with a matching grant program, which
would mean that they would be more likely to be targeting exactly
the areas that are most in need, because their dollars would be invested, as well.
I thank you for the opportunity to have addressed you. I appreciate the fact that you have us here today. And I hope that all of
us move forward in a cooperative way to try to do what we can for
those areas of America that need to be served both by telephone
service and by advanced services.
Thank you.
[The prepared statement of Mr. Burke follows:]
PREPARED STATEMENT OF HON. JOHN DOWNES BURKE, BOARD MEMBER, VERMONT
PUBLIC SERVICE BOARD; MEMBER, FEDERAL-STATE JOINT BOARD ON SEPARATIONS;
AND MEMBER, FEDERAL-STATE JOINT BOARD ON UNIVERSAL SERVICE
I. Introduction
I thank the Committee for the invitation to speak today. Federal Universal Service policy is of great importance to the Nation, and particularly to states, like
Vermont, where it is expensive to provide telephone service.
To introduce myself, I have been a Member of the Vermont Public Service Board
for 6 years. I have served on the Federal-State Joint Board on Separations since
2003. Last year I was also appointed as a Member of the Federal-State Joint Board
on Universal Service. With Commissioners Baum of Oregon and Landis of Indiana,
I have also served as one of five state NARUC Commissioners who oversaw the industrys development of the current Missoula Plan.
II. The Statute
The existing Universal Service law, Section 254, was passed in 1996. It was a significant step forward in establishing universal availability of telephone services in
this country. Section 254 of the Act repeatedly imposes the duty on both the FCC
and the Universal Service Joint Board to preserve and advance universal service.1
1 See

VerDate Nov 24 2008

14:09 Mar 21, 2012

47 U.S.C. 254(b), 254(b)(4), 254(b)(5), 254(d), 254(f).

Jkt 073387

PO 00000

Frm 00020

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

17
More specifically, the statute lists six goals, some of which apply primarily to distribution of Universal Service support, and some of which apply to collection. Notable in this list is subdivision (3), which requires that rural access to telecommunications and information services, including advanced telecommunications and information services be reasonably comparable to those services provided in urban
areas. It also requires that the rural rates charged be reasonably comparable to
rates charged for similar services in urban areas.
Reasonably comparable rates is the heart of the high-cost support system, and the
courts have taken this goal seriously. Twice, the Court of Appeals in Denver has
remanded FCC decisions because the FCC had not shown how its programs satisfy
that goal.2 Developing a system in compliance with Section 254 should be an important priority for the Joint Board and the FCC.
A. ChallengeCompetition and the Growth of CETCs
The most urgent problem for Universal Service is the rapid growth of funding for
competitive telecommunications carriers. There is nothing inherently wrong with
providing support to competitors, but our current policy is on a self-destructive path
that could jeopardize the entire Universal Service system, and it is my opinion that
subsidizing robust competition was never an underlying goal of this Fund.
Our current policy was adopted to promote competitive neutrality, a seventh
principle that the first Joint Board added to the list of goals for universal service.
As we have applied it, this principle has led to the equal payment rule. Under
this rule, carrier A who is an Incumbent Eligible Telecommunications Carrier
(IETC), and carrier B, who is a Competitive Eligible Telecommunications Carrier
(CETC), receive equal support per line. This was seen as neutral, even though only
carrier A must submit its costs.
This equal payment rule was originally conceived as a way to transfer support
when a CETC wins a customer from an IETC. Review of the record shows that the
Joint Board did not anticipate that households with one carrier and one telephone
line would begin to have two or more carriers and multiple cellphone lines in addition to the classical Plain Old Telephone Service (POTS) line. Today, each of these
lines may draw a quota of Universal Service support.
The equal payment rule never acknowledged the effects of economies of scale, one
of the basic characteristics of networks. When two carriers divide a market that previously was served by one, the total cost of serving that area can go up, not down,
particularly if the area served is high-cost and rural. As Chairman Martin has repeatedly pointed out, our current policy has the effect of supporting construction of
multiple networks in areas where constructing the first network has been very expensive. This has understandably produced explosive growth in the support provided to Competitive ETCs. As Commissioner Tates slides show, this support has
been growing at 101 percent per year for the last 4 years, and is approaching $1
billion. Moreover, the number of new CETC applications suggests the growth will
continue into next year and beyond. Even though support to ILECs has held fairly
level during this period, rapid CETC growth creates risk for the entire Universal
Service mechanism.
I have recommended an immediate CETC cap for all carriers whose support depends on the equal payment rule. I recommended that the cap apply by study area,
so that areas without CETC support would remain that way until a new CETC support system is devised. In areas where there is already some CETC support, that
amount would be divided among the competitive carriers that obtain designations.
Over the longer term, it is imperative that we develop clearer policies about how
we expect existing networks to be supported in high-cost areas, how many networks
we are willing to support, and how they will be selected.
B. ChallengeUneven Support
One of the earliest decisions made after the 1996 Act passed was to create separate tracks for the Universal Service provided to rural and so-called nonrural
carriers. That decision has continued to this day.
Today, rural and nonrural carriers have largely distinct support systems. The
mechanisms differ in many significant ways, but the overall effect is that support
for larger nonrural carriers is significantly less than support for smaller rural carriers. Today the average rural carrier receives $13.68 per line per month in highcost support. The average nonrural customer receives 66 cents per line of high-cost
support, and most of that goes for interstate cost and not for local rate reductions.
In sum, customers of large carriers receive about five cents of high-cost support for
2 Qwest

VerDate Nov 24 2008

14:09 Mar 21, 2012

Comm. Intl Inc. v. FCC, 398 F.3d 1222, 1235 (10th Cir. 2005).

Jkt 073387

PO 00000

Frm 00021

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

18
every dollar paid to benefit the customer of rural companies, not due to differences
in need, but rather, to the size of the company that serves them.
This would be fine if nonrural carriers had no rural customers. In fact, the match
between rural carrier and rural customer works fairly well in the Midwest where
there are hundreds of rural companies. But the equation between rural carrier
and rural customer does not work well in New England or in the Appalachian region, where Bell companies still serve large rural areas.3 In truth, millions of rural
customers are served by larger carriers. Among the so-called nonrural companies,
more than one customer in five is actually a rural customer.4
The problem bites most deeply in states, like Maine, Montana, Wyoming and
Vermont, where there are no large cities that can subsidize rural areas through retail rate averaging. These states suffer from a double disability: the absence of large
cities eliminates the possibility of averaging high and low cost areas to develop
lower average rates overall; and the absence of smaller rural telephone companies
reduces the support available to rural customers.
The disparity between rural and nonrural companies has only become worse over
time. Rural customers served by large companies today are not only likely to have
higher rates, they probably have less access to broadband as well.
C. ChallengeBroadband
Broadband is probably the most important current challenge for universal service.
Section 254 directs that access to advanced services should be provided in all regions of the Nation. Yet many states have large areas where broadband is available
only by satellite. It has been widely reported that the United States is falling behind, year by year, in the percentage of our citizens who buy broadband.
The Joint Board should give serious consideration to adding broadband to the official list of supported services. Section 253 gives us detailed guidance for this decision. The statute recognizes that Universal Service is an evolving level of telecommunications services. We must consider whether such broadband telecommunications services are essential to education, public health, or public safety and
whether a substantial majority of residential customers have actually subscribed.5
As Consumer Advocate Gregg has pointed out to us, a majority of residential customers may soon actually subscribe to broadband.
One possible problem is that Section 254 allows us to add only telecommunications services to the existing list of supported services. The FCC has declared
that several kinds of broadband Internet services are actually information service,
not telecommunications service. So, even though Section 254 tells us explicitly
that access to advanced telecommunications and information services should be
provided in all regions of the Nation, we will need to examine carefully whether
these FCC rulings bar use of Section 254 as a vehicle to promote broadband.
A second possible problem is that including broadband in the definition of Universal Service could inadvertently disqualify some existing carriers who provide
POTS or Plain Old Telephone Service. Section 254 does not specifically anticipate allowing funding for services that do not meet the minimum requirements for
eligibility. We would need to move carefully to allow existing carriers a reasonable
transition period to meet any new requirements.
Another concern is that to include broadband in the definition of Universal Service could greatly expand the size of the high-cost fund. This is a serious concern,
but we should not assume that broadband services will be supported in the same
ways that we now support POTS.
The Joint Board has recently sought comment on the use of auctions, and we are
examining the potential for newer technologies to better target existing support. I
believe that we should also examine matching grants. Many Federal agencies, from
Transportation to Education, today promote good state policy through the use of
such matching grants. If applied to broadband, a system of matching grants could
easily be controlled fiscally by implementing an annual funding cap. Also, matching
grants would be most likely to be effective. States generally know the most about
their own broadband needs, and a mechanism that required a state matching share
would be very likely to focus support in areas where a problem really exists.
Earlier this winter, Vermont Governor James Douglas outlined to our state legislature an initiative that would authorize state bonding to provide broadband in
3 The converse problem also exists. A few so-called rural carriers actually serve low cost suburbs.
4 Vicki M. Hobbs, and John Blodgett, The Rural Differential: An Analysis of Population Demographics in Areas Served by Rural Telephone Companies, Rural Policy Research Institute, 1999
at 2 (21 percent of large carrier customers are rural, based upon 1990 census).
5 See 47 U.S.C. 254(c)(1).

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00022

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

19
unserved areas. A Federal matching grant for broadband deployment would allow
us to stretch our limited state dollars. It would greatly assist Vermont and other
states that are still struggling to provide a first broadband connection to many of
their citizens.
I also agree with Commissioner Copps that data quality is a problem for
broadband. Data indicating which Zip Codes have broadband is misleading. Knowing that broadband is available somewhere within a zip code is little solace to an
individual customer who cant buy it from anyone. The Joint Board should be collecting data on broadband at a much finer scale than it does now, and the technology clearly exists to do this.
D. ChallengeLimiting Fund Size
I have mentioned the need to equalize support for all rural customers and my desire to expand support to broadband. I also want to emphasize that a rational Universal Service policy can achieve these goals without unduly increasing the size of
the national Fund, possibly without increasing it at all.
The existing Universal Service system has not been designed as a single system.
Rather, it is a series of eight separate programs that were created incrementally
over two decades.6 A few programs have been modified, but none has ever been replaced. Each new program typically focused on some cost component or company
characteristic that seemed relevant at the time.7 But we have never taken a comprehensive and multi-jurisdictional view of carrier costs, and we have never replaced even one older program.
Another problem with the existing system is that it provides the most support to
the smallest companies, not necessarily those with the highest costs. The most obvious example today is Local Switching Support, which does not even attempt to limit
support to carriers with high costs.8
If we could design a comprehensive system, we could adopt a single definition of
total cost, and we could find new efficiencies by eliminating support to carriers that
do not have high overall costs.
As I mentioned above, matching grants can be another tool to maintain fiscal discipline. Federal matching programs in other policy areas routinely live within their
budgets.
For these reasons, I believe that the existing fund size could be reduced, or we
could broaden the scope of the Fund to cover broadband, without unduly harming
rate payers and without violating any of the principles contained in Section 254(b).
E. ChallengeIntercarrier Compensation and Separations
I mentioned above that I have been privileged to serve on both Joint Boards and
the NARUC Intercarrier Compensation project. This has convinced me that Universal Service is intimately tied both to separations and to intercarrier compensation.
6 The first Universal Service program was the High Cost Loop program, and was created in
1984.
7 For example, the High Cost Loop program addressed a 1984 change in the jurisdictional separation of loop costs. Today this program exceeds $1 billion, but there is no similar program
for the interoffice transport costs of rural companies, costs that for some companies can be even
larger than loop costs. Likewise, the Interstate Access Reform and Interstate Common Line Support programs were created in 2000 and 2001 to replace revenues lost through reform of interstate toll access rates.
8 This support mechanism was created in 1987 when the FCC made a change affecting the
separation of costs affecting the cost recovery for local class 5 switches. Originally known as
DEM weighting, this mechanism allowed ILECs with 50,000 or less access lines to allocate
a higher percentage of their local switching costs to the interstate jurisdiction. The greatest benefit went to ILECs that already had the largest interstate usage and to the ILECs that had
the fewest lines, according to the following table.
Number of Access Lines
Weighting for Interstate Dial
in Study Area
Equipment Minutes Separations Factor

0 to 10,000
3.0

10,001 to 20,000
2.5

20,001 to 50,000
2.0

50,001 or more
1.0
See 47 C.F.R. 36.125(f ). While the FCCs rules for this program no longer explicitly differentiate based upon size, the programs 1996 support parameters were frozen in place by a reformulation that took effect on January 1, 1998, thereby indefinitely perpetuating the size-based distinction. See 47 C.F.R. 36.125(f ).

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00023

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

20
Separations has had a particularly close historical relationship to universal service. Before 1996, Universal Service programs were enacted in the form of separations rules. Although these programs were designed to reduce or avoid an increase
in local rates, they acted through separations rules and created inter-jurisdictional
cost transfers that ultimately raised interstate access and toll rates.9 Even more recent programs, like the High Cost Modeling Program that applies to larger carriers,
rely on separations factors to avoid the double-recovery of costs that have been separated to the interstate jurisdiction.10
Universal service also has a close historical relationship to intercarrier compensation. Several Universal Service programs were created solely as components in
intercarrier compensation reforms. For example, the CALLS program, adopted by
the FCC in May of 2000, reformed interstate access rates for large price cap carriers. The following year, 2001, the Commission adopted the MAG order that did
essentially the same thing for smaller rate-of-return carriers. Each order created
a new Universal Service mechanism. This year those programsInterstate Access
Support and Interstate Common Line Supportwill cost $1.9 billion.
The current version of the Missoula Plan, now pending before the Commission,
would add another layer. It proposes additional FCC payments of $2.5 billion to finance the reform of new kinds of intercarrier payments, such as intrastate access
and reciprocal compensation.
The close interaction among these programs shows why two Joint Boards sometimes find it difficult to identify comprehensive solutions. One can seldom make a
recommendation on any of the three topics without affecting the other two. Perhaps
Congress should consider a new and more comprehensive mechanism for cooperation
between the FCC and the states, particularly in policy areas requiring coordination
of rates and cost assignments.
F. Improving the Uses of USF Dollars
Some carriers have criticized the existing support mechanisms for being insufficiently specific geographically. I agree that more detailed targeting of support for
competitive carriers could possibly increase their investment in underserved areas.
However, I think the Joint Board and the states have adequate tools now to address
this issue. The Joint Board is looking at proposals from industry that would mandate greater disaggregation of existing support, with this result in mind.
We should not forget that states already have some tools to encourage carriers
to invest in unserved areas. States annually must certify the proper use of Universal Service support. These certifications offer states a chance to review where
Federal funds have been spent, and some states have required detailed investment
plans as a condition of annual certification.11

The CHAIRMAN. I thank you very much, Mr. Burke.


And now, Im pleased to yield to my colleague, Senator Rockefeller.
STATEMENT OF HON. JOHN D. ROCKEFELLER IV,
U.S. SENATOR FROM WEST VIRGINIA

Senator ROCKEFELLER. Thank you, Mr. Chairman.


Im very pleased to welcome Bill Jack Gregg back again before
the Committee. He has served, with the distinction that is his, as
the Director of the Consumer Advocate Division of the West Virginia Public Service Commission, since the office was created by a
particularly brilliant Governor who happened to be presiding at
that time.
[Laughter.]
The CHAIRMAN. You talk about yourself?
9 The High Cost Loop (1984) and DEM Weighting (1987) programs were both codified in separations rules. Inter-jurisdictional cost transfers still exist. See, e.g., 47 C.F.R. 36.603 which describes the High Cost Loop program as a loop cost expense adjustment.
10 See 47 C.F.R. 54.309 (support equals 76 percent of difference between cost and benchmark).
11 Vermont also has used another tool, the designation process, to promote rural investment.
Vermonts sole CETC has a designation as ETC that expires from time to time. Before the designation is extended, the Public Service Board reviews the carriers investment history and the
geographic areas to which it has extended service.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00024

Fmt 6633

Sfmt 6601

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

21
Senator ROCKEFELLER. Of course.
Over the years, hes providedwell, look, hes a visionary as to
how this whole thing ought to work. Hes spent his life on it. Were
lucky to have him before us.
Welcome, sir.
STATEMENT OF HON. BILLY JACK GREGG, DIRECTOR,
CONSUMER ADVOCATE DIVISION,
PUBLIC SERVICE COMMISSION OF WEST VIRGINIA

Mr. GREGG. Thank you. Thank you, Mr. Chairman, Mr. Vice
Chairman, members of the Committee.
As youve heard from the other speakers before me, there is
amazing agreement amongst the members of the Joint Board as to
what the current problems are with the high-cost fund. The current
system is unsustainable. Its inconsistent, and incredibly complex.
Its growing out of control. Its distributed poorly.
The sad fact is that the advent of competition into telecommunications has actually caused a substantial increase in the size of the
high-cost fund, as youve seen visually on the slides presented by
Commissioner Tate. This has increased the burden on all consumers, and it didnt have to be this way.
As noted in both the House and Senate reports on the Telecommunications Act of 1996, competition was supposed to lower the
cost of Universal Service as providers competed for the Universal
Service subsidy. The FCC initially kept true to the intent of the
Act. Universal Service high-cost support, as modified by the Commission for the advent of competition, was a technologically and
competitively neutral zero-sum game. That is, the Universal Service subsidy was portable to whichever eligible telecommunications
carrier won the customer. The ETC gaining the customer won the
subsidy. The ETC losing the customer lost the subsidy. The Commissions approach was upheld by the Fifth Circuit Court of Appeals in the case of Alenco Communications versus the FCC in
2000.
Unfortunately, without explanation, the Commission abandoned
its rulemaking proceeding to define captured and new lines and
deleted a section of its rules which had reduced support to an incumbent when a competitive ETC won a customer. As a result, the
Commission began providing support for all lines of all ETCs serving high-cost areas.
In 1999, this did not seem like a big deal. Unfortunately, the unforeseen consequences of these actions have been dramatic. By deciding to support all lines of all ETCs in high-cost areas, the Commission opened the door to supporting multiple wireless networks
in high-cost areas, which supplied supplementary, rather than substitute, services. Far from being a zero-sum game in which ETCs
compete for the Universal Service subsidy while the size of the
Fund stays relatively the same, the current system is a no-losers
support system in which all ETCs receive support for all lines they
serve in high-cost areas, no matter how duplicative or costly this
additional support may be.
Under the current system, far more than affordable access to the
telecommunications network is being provided. The high-cost fund
now provides support to multiple networks in high-cost areas,

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00025

Fmt 6633

Sfmt 6601

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

22
where, previously, none had been able to exist without an explicit
subsidy. The current system of providing support to all lines of all
ETCs in high-cost areas must be ended if we are to have a rational
and sustainable high-cost support system. In fact, when multiple
providers are able to offer service within the same area, it raises
the question of whether that area should continue to receive highcost support at all. Because of the complex, disparate, and often
unrelated bases of the different high-cost support mechanisms, and
the rapidly escalating size of the high-cost fund caused by increasing payments to competitive ETCs, the Joint Board has begun to
look at new alternatives to bring rationality back to the high-cost
fund, as you have heard this morning.
Unfortunately, while we contemplate these proposals, the Fund
will continue to grow to an ever more unsustainable size. The highcost fund has increased by a billion dollars, as youve heard, over
the past 3 years, driven by new payments to competitive ETCs. In
addition, the FCC currently has before it pending over 30 applications for ETC status from wireless carriers, including two from
Cingular for the states of Virginia and Georgia. Cingular is the
largest wireless provider in the United States. The FCC has estimated that, if it grants all of the ETC applications pending before
it today, the high-cost fund will rise to five and a half billion dollars by 2009. And if Cingular continues to seek ETC status,
Verizon Wireless, the second largest wireless provider, will be
forced to follow suit. The result will be a high-cost fund surpassing
$6 billion and approaching $7 billion. A fund of this size will not
only impose unacceptable burdens on American consumers, it will
also severely limit our ability to add new services, such as
broadband, to the list of services supported by Universal Service.
In order to be stable and sustainable in the long-term, the Universal Service Fund must be configured like a pyramid. It must
have a broad and stable base of contributions at the bottom and
a narrow, but sufficient, distribution of support at the top. The current Universal Service Fund requires work on both ends of the
structure. Issues related to the contribution base must be resolved.
Since all benefit from Universal Service, all should contribute. In
addition, the limited resources of the Fund must be properly distributed and targeted to carry out the purposes of the Act. In order
to continue the public policy success of the Universal Service Fund,
we must support access, not excess.
Thank you very much.
[The prepared statement of Mr. Gregg follows:]
PREPARED STATEMENT OF HON. BILLY JACK GREGG, DIRECTOR, CONSUMER
ADVOCATE DIVISION, PUBLIC SERVICE COMMISSION OF WEST VIRGINIA
My name is Billy Jack Gregg and I am the Director of the West Virginia Consumer Advocate Division. My office is charged with the responsibility of representing West Virginia utility ratepayers in state and Federal proceedings which
may affect rates for electricity, gas, telephone and water service. My office is also
a Member of the National Association of State Utility Consumer Advocates
(NASUCA), an organization of 43 state utility consumer advocate offices from 41
states and the District of Columbia, charged by their respective state statutes with
representing utility consumers before state and Federal utility commissions and be-

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00026

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

23
fore state and Federal courts.1 I am a former member of the Board of Directors of
the Universal Service Administrative Company (USAC) and the Rural Task Force,
and have served on the Federal-State Joint Board on Universal Service since March
2002. I greatly appreciate the opportunity to testify at this legislative hearing on
the challenges currently facing the Federal Universal Service Fund (USF or the
Fund).
I. Background
The most important issue facing the Federal USF today is adapting the Fund to
a competitive environment and ensuring its long-term sustainability. As the telecommunications market changes rapidly, we must ensure that the USF is sufficient,
predictable and affordable for all parties involved: fund recipients, telecommunications providers and consumers. Before I address the current problems facing the
USF, I believe it is appropriate to review the Funds achievements since the passage
of the Telecommunications Act of 1996 (the Act).
The nations commitment to Universal Service was codified in Section 254 of the
Act. The purpose of Section 254 was to ensure that all Americans have access to
affordable, quality telecommunications services.2 Based upon the requirements of
Section 254, the FCC, after consultation with the Federal-State Joint Board on Universal Service, created a new Federal USF in 1997 containing several distinct support mechanisms. Total USF funding has grown from $1.8 billion in 1997 to approximately $7.2 billion during 2007. While these support amounts are large, they must
be kept in perspective. Total telecommunications revenues in the United States last
year were in excess of $230 billion. By annually collecting and redistributing approximately 3 percent of these total revenues, we are able to: provide affordable access to phone service in all high-cost areas of the nation; support low-income customers; assist rural healthcare providers; and connect all classrooms to the Internet.
Moreover, all states and territories benefit from the USF as shown on Attachments
1 and 2.3 Thats quite an accomplishment, and one that everyone involved in the
USF should be proud of as we move forward to ensure the long-term sustainability
of the Fund.
However, as with all things, somebody must pay for the Funds benefits. That
somebody is the American telecommunications consumer in every state and territory. Although all states benefit from the USF, some states pay far more into the
Fund than they receive back in support, as shown on Attachments 3 and 4.4 The
concept of sustainability encompasses both the size of the Fund and the relative
burden it imposes. In order to ensure that the USF is sustainable for the long-term,
we must ensure that the USF remains affordable for the individual consumer and
for the payer states. As I will discuss in detail later, the biggest threat to the longterm sustainability of the USF is the burden imposed by the unrestrained growth
of the High Cost Fund.
II. The Long Term Sustainability of the Universal Service Fund
As previously mentioned, the Federal USF has grown from $1.8 billion to $7.2 billion since the Act was passed. During this same time the USF assessment factor,
which is paid by all local, long distance and wireless customers in the United States
based on interstate revenues, has more than doubled, from less than 5 percent to
over 11 percent.5 Almost everyone who addresses the issue of the long-term sustainability of the USF has the same prescription: broaden the contribution base and
properly control the distribution of funds from the USF. However, depending on the
1 NASUCA has the unique position of representing consumers in states which benefit from
universal service, as well as consumers who must pay the cost of Universal Service. In most
respects, my testimony reflects the positions taken by NASUCA, although there are some areas
where NASUCA has not yet reached a consensus position.
2 Section 254 of the Act enshrined and expanded Universal Service principles which had been
followed by the FCC for decades.
3 Attachments 1 and 2 show actual disbursements to states during 2005 under each of the Federal USF support mechanisms. Attachment 1 ranks the states based on total support received.
Attachment 2 considers the number of access lines in each state, and ranks the states based
on monthly support received per line.
4 Attachments 3 and 4 show the same disbursements as Attachments 1 and 2, but also include
the USF payments made by consumers in each state during 2005. Attachment 3 ranks the states
based on total net support received, while Attachment 4 ranks the states on net per line support
received. Negative numbers indicate that states paid more in USF assessments than they received in USF benefits.
5 The assessment factor was 9.7 percent during the first quarter of 2007 and is expected to
rise above 11 percent for the second quarter.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00027

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

24
interest group making the recommendation, the actual method of broadening the
base and controlling the distribution of funds can vary wildly.
The FCC and Congress have wrestled with the issue of the funding base for over
4 years. Although numerous ideas and proposals to broaden the contribution base
have been brought forth, none have been implemented. Many parties oppose broadening the contribution base on the grounds that it will only lead to more profligate
spending of money paid into the USF. I am firmly convinced that unless we first
bring the distribution of the High Cost Fund under control, no progress will be
made on the contribution side.
In looking at the long-term sustainability of the Fund, we need to review the status of funds paid out by the individual support mechanisms which make up the
overall USF. A quick review of the four funds making up the Federal USFthe
High Cost Fund, the Low Income Fund, the Schools and Libraries Fund, and the
Rural Health Care Fundshows that the High Cost Fund is the most problematic.
Set forth below are the collections for each of these funds in 2003 and projected for
2007.6

Change in USF Funding Mechanisms


[20032007]
$ Millions
USF Fund

2003

High Cost Fund


Low Income Fund
Schools & Libraries Fund
Rural Health Care Fund
Total

2007

Change

3,261.1
712.9
2,184.0
27.9

4,270.8
766.8
1,988.5
160.0

1,009.7
53.9
195.5
132.1

6,185.9

7,186.1

1,000.2

As can be seen, the High Cost Fund has grown by over a billion dollars since
2003, while the other funds have shown modest or negative growth in the same period. The Schools and Libraries Fund has been capped at $2.25 billion a year since
its inception. The Rural Health Care Fund has likewise been capped at $400 million
a year, although annual expenditures have come nowhere near that level. The Low
Income Fund has been the focus of repeated state and Federal efforts to increase
participation, yet funding has not grown substantially over the past 4 years. The
High Cost Fund is clearly the main driver in the growth in the overall Fund and
the USF contribution factor.
Within the High Cost Fund, support for competitive eligible telecommunications
carriers (ETCs), and more particularly wireless carriers, has been the sole cause of
growth since 2003. As shown below, payments to competitive ETCs have soared
from $126.7 million in 2003 to $1.2 billion projected for 2007.7

6 The 2007 figures are based on USAC demand projections for the first two quarters, with
funding for the third and fourth quarter assumed to be the same as in the second quarter. A
graphic display of the growth of each of the funds since 2000 is set forth on Attachment 5.
7 Once again, the 2007 figures for CETCs are based on USAC projections for the first two
quarters of 2007, with funding for CETCs for the third and fourth quarters assumed to be the
same as the second quarter.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00028

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

25

While USF support payments to competitive ETCs have increased tenfold, payments to both rural and non-rural incumbent ETCs have actually declined, as
shown below.8

Change in Funding to ETCs


[20032007]
$ Millions
ETCS

2003

Rural Incumbents
Non-rural Incumbents
Competitive ETCs
Total

2007

Change

2,467.0
767.9
126.7

2,415.5
689.8
1,220.2

51.5
78.1
1,093.5

3,361.6

4,325.5

963.9

Payments to rural incumbents have been held in check by a cap on the High Cost
Loop Fund. This cap does not apply to competitive ETCs. Payments to non-rural incumbents have been limited by loss of lines and a ceiling on the Interstate Access
Support Fund.
It should not be surprising that funding for competitive ETCs has increased. After
all, before the advent of competition incumbents received 100 percent of high-cost
funding. It was expected that as competitors gained ETC status and won customers
in high-cost areas, their high-cost funding would rise. What is surprising is that incumbent support has not dropped by an amount proportionate to the increase in
competitive ETC funding. In other words, the advent of competition has actually

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00029

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

301gregg1.eps

8 The totals shown in the table differ slightly from the High Cost Fund totals shown in the
table on page 24 because they are not adjusted by interest earnings, administrative costs and
out-of-period adjustments.

26
caused a substantial increase in the size of the High Cost Fund, and increased the
burden on all consumers. It did not have to be this way.
III. Competition and the Universal Service Fund
It has often been said that the twin pillars of the 1996 Telecommunications Act
were competition and universal service. Competition would allow consumers to enjoy
lower prices and better services, while Universal Service would ensure that all
Americans, even those in rural and high-cost areas, would share in the benefits. Not
only was the introduction of competition expected to lower prices of telecommunications services, it was supposed to lower the cost of Universal Service as providers
competed for the Universal Service subsidy. As the House and Senate Reports on
the Act stated:
. . . as the current system of internal and external subsidies is replaced by a
system consisting primarily of external subsidies, the total amount of subsidies
collected from low-cost customers and passed on to high-cost customers would
not change significantly. Over time, CBO [Congressional Budget Office] expects
that the operating costs of telephone companies would tend to fall as a result
of competitive pressures and the total amount of subsidies necessary would decline.9
. . . competition and new technologies will greatly reduce the actual cost of providing Universal Service over time, thus reducing or eliminating the need for
Universal Service support mechanisms as actual costs drop to a level that is at
or below the affordable rate for such service in an area. . . .10
This view was echoed by Senator Stevens during debate on the Act:
[The Act] opens up the local market to competition while still preserving the
concept of universal service. It does so by taking advantage of new technologies
which are intended to reduce the cost of all services, including universal service.
In fact, I find it interesting that the Congressional Budget Office has said that
this bill will reduce the cost of Universal Service from the existing system by
at least $3 billion over the next 5 years.11
The High Cost Fund began in a monopoly environment prior to the passage of
the Act. Since 1996 the FCC has struggled to adapt the USF to a competitive environment where multiple providers could offer the same or similar services to consumers. In implementing the Universal Service provisions of the Act, the FCC initially kept true to the Acts intent. In the First Report and Order on Universal Service, the Commission described its overall approach to universal service:
. . . Universal Service will be sustainable in a competitive environment; this
means both that the system of support must be competitively neutral and permanent, and that all support must be targeted as well as portable among eligible telecommunications carriers. . . . By following the principle of competitive
neutrality, we will avoid limiting providers of Universal Service to modes of delivering that service that are obsolete or not cost effective.12
The Commission also dealt directly with the issue of which ETC would receive
high-cost support:
We adopt the Joint Boards recommendation to make rural carriers support
payments portable. . . . [A] CLEC [competitive local exchange carrier] that
qualifies as an eligible telecommunications carrier shall receive Universal Service support to the extent that it captures subscribers formerly served by carriers
receiving support based on the modified existing support mechanisms or adds
new customers in the ILECs study area. We conclude that paying the support
to a competitive eligible telecommunications carrier that wins the customer or
adds a new subscriber would aid entry of competition in rural areas. [Emphasis
added.] 13
In short, Universal Service high-cost support, as modified by the Commission for
the advent of competition, was a technologically and competitively neutral zero sum
9 House Report No. 104204(I) (1995), Arnold & Porter Legislative History Pub. L. 104104
(A&P) at 60.
10 Senate Report No. 10423, A&P at 254 (1995).
11 141 Congressional Record S7881 (1995), A&P at 210.
12 In re: Federal-State Joint Board on Universal Service, CC Docket No. 9645, Report &
Order (May 8, 1997); as corrected by Erratum, FCC 97157 (June 4, 1997) at 19 & 49; affd
in relevant part sub nom. Texas Office of Public Utility Counsel v. FCC, 183 F.3d 393 (5th Cir.
1999). This order will be referred to as the First Report & Order.
13 First Report & Order, 311. See also 287289; 312.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00030

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

27
game: the Universal Service subsidy was portable to whichever ETC won the customer. The ETC gaining the customer won the subsidy, the ETC losing the customer
lost the subsidy. As part of this framework, the Commission revised its rules to add
Section 54.307(a)(4) which stated:
The amount of Universal Service support provided to such incumbent local exchange carrier shall be reduced by an amount equal to the amount provided to
such competitive eligible telecommunications carrier.14
The Commission stated that this rule change was necessary to ensure that when
a competitive ETC received support for a customer, . . . the incumbent LEC will
lose the support it previously received that was attributable to that customer. 15
The Commissions approach was upheld by the Fifth Circuit Court of Appeals in the
case of Alenco Communications, Inc. v. FCC:
The FCC must see to it that both Universal Service and local competition are
realized; one cannot be sacrificed in favor of the other. The Commission therefore is responsible for making the changes necessary to its Universal Service
program to ensure that it survives in the new world of competition. . . .
. . . [T]he [FCCs universal service] order provides that the Universal Service
subsidy be portable so that it moves with the customer, rather than stay with
the incumbent LEC, whenever the customer makes the decision to switch local
service providers. . . . The purpose of Universal Service is to benefit the customer, not the carrier. Sufficient funding of the customers right to adequate
telephone service can be achieved regardless of which carrier ultimately receives
the subsidy.16 [Emphasis in original.]
Unfortunately, in November 1999, without explanation the Commission abandoned its rulemaking proceeding to define captured and new lines and deleted Section 54.307(a)(4) of its rules which had reduced support to an incumbent when a
competitive ETC won a customer.17 Finally, in April 2000, the Commission effectively abandoned the distinction between new, captured, and other lines served
by ETCs, stating . . . a competitive eligible telecommunications carrier receives
support for each line it serves based on the support the incumbent local exchange
carrier would receive for serving the line. 18
The unforeseen consequences of these actions have been dramatic. By deciding to
support all lines of all ETCs in high-cost areas, the Commission opened the door
to supporting multiple wireless networks which supplied supplementary, rather
than substitute services. As previously discussed, this supplementary support to
wireless ETCs has added a billion dollars to the High Cost Fund since 2003.19 Far
from being a zero sum game in which ETCs compete for customers while the size
of the Fund stays relatively the same, the current system is a no losers support
system in which all ETCs receive support for all lines they serve in high-cost areas,
no matter how duplicative or costly this additional support may be.20
Under the current system, far more than affordable access to the telecommunications network is being provided. The High Cost Fund now provides support to
multiple networks in high-cost areas, where previously none had been able to exist
without a subsidy. If a customer in a high-cost area receives two landlines from the
incumbent wireline ETC, and three wireless phones from a competitive ETC, all of
these lines receive high-cost support. Even more bizarre, if the rural incumbent ETC
actually loses lines, support for both the incumbent ETC and the competitive ETC
will go up as a result of the equal support rule.21 The result has been a rapid esca14 In re: Federal-State Joint Board on Universal Service, CC Docket No. 9645, Fourth Order
on Reconsideration (Dec. 30, 1997) at 84; App. A, Item 6, 47 C.F.R. 54.307(a)(4).
15 Id.
16 Alenco Communications, Inc. v. FCC, 201 F.3d 608, 615 & 621 (5th Cir. 2000).
17 In re: Federal-State Joint Board on Universal Service, CC Docket No. 9645, Ninth Report
& Order and Eighteenth Order on Reconsideration (Nov. 2, 1999), at 90; App. C, Item 7.
18 In re: Federal-State Joint Board on Universal Service, CC Docket No. 9645, Order (April
7, 2000), at 16.
19 In essence, the USF has created a $1 billion wireless infrastructure fund. This was done
without any explicit policy decision or directive by either the Congress or the Commission. It
just happened, based on the incentives created by the high-cost support rules.
20 The fact that multiple providers are able to offer service within a particular area raises the
question of whether that area should properly be able to receive continued high-cost support.
21 The equal support rule, found in 47 C.F.R. 54.307(a)(1), provides that a competitive ETC
will receive per line support equal to the support received by the incumbent ETC. Because the
High Cost Loop mechanism is designed to recover an incumbents full revenue requirement reContinued

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00031

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

28
lation of support as competitive ETCs have rushed in to take advantage of the rules
created by the FCC.
One outrageous example of the current system is found in the AT&T (BellSouth)
service territory in Mississippi. AT&T as the incumbent non-rural carrier receives
$101.2 million in High Cost Support annually. In addition, there are sixteen (16)
other competitive ETCs receiving $118.5 million in High Cost Support annually for
providing service in the same study area.22 Most of this CETC support goes to wireless ETCs, including $59.1 million to AT&Ts wireless subsidiary, Cingular. While
there is no doubt that Mississippi is a high-cost area, the Acts requirement to provide affordable access does not require providing subsidies to multiple networks
serving the same customers. The current system of providing support to all lines
of all ETCs in high-cost areas must be ended if we are to have rational and sustainable high-cost support system.
Because of the complex, disparate and often unrelated bases of the different highcost support mechanisms, and the rapidly escalating size of the High Cost Fund
caused by increasing payments to competitive ETCs, the Joint Board has begun to
look at new alternatives to bring rationality back to the High Cost Fund. One of
these proposals is reverse auctions; another is newer, more sophisticated modeling
and more precise targeting of support based on new mapping technology. These proposals will need much work before it is determined if they are ready to be implemented on a national or even a pilot project scale.
Unfortunately, while we contemplate these proposals, the Fund will continue to
grow to an ever more unsustainable size. The High Cost Fund has increased by $1
billion over the past 3 years driven by new payments to competitive ETCs. In addition, the FCC currently has pending before it over thirty (30) applications for ETC
status from wireless carriers, including two from Cingular for the states of Virginia
and Georgia. The FCC has estimated that if it grants all of the ETC applications
pending today, the High Cost Fund will rise to $5.5 billion by 2009. If Cingular, the
largest wireless carrier, continues to seek ETC status, Verizon Wireless, the second
largest, will be forced to follow suit. The result will be a High Cost Fund surpassing
$6 billion and approaching $7 billion. A fund of this size will not only impose unacceptable burdens on American consumers, but will severely limit our ability to add
new services, such as broadband, to the list of services supported by universal service.
As a result, the Joint Board is currently considering several proposals to cap the
High Cost Fund while we consider long-term solutions on how to adapt the Universal Service system to the new competitive environment by properly targeting
support and ensuring that the Fund does not grow to an unsustainable size. In fact,
one of the difficulties confronting policymakers in this area is the lack of any upward limit on the Fund expressed by Congress. It is interesting to note that in the
currently pending S. 101, the Universal Service for Americans Act, Section 202 creates a $500 million a year Broadband for Unserved Areas Program. This is similar
to funding under the existing cap on the Schools and Libraries Fund. Moreover, Section 202 makes clear that distributions from the Broadband fund may only be made
to one facilities-based broadband provider in each unserved area. Based on the
wording of Section 202, policymakers know exactly how much they have to spend,
and can then attend to the issues of how to equitably distribute the Fund in accordance with the principles established by Congress. While a limitless Universal Service Fund may have made sense when we were faced with making previous implicit
subsidies explicit, eleven years after the passage of the Act it may be time for Congress to also express its opinion on the ultimate size of the High Cost Fund.
IV. The Contribution Base
Ensuring the long-term sustainability of the Fund will require not only controlling
the size and distribution of the fund, but also broadening the contribution base.
Moreover, until the distribution and sizing issues are solved, it is not likely that a
consensus will develop concerning how to address the contribution base.
The funding base for the USF has not kept pace with the growth in the fund, resulting in higher and higher USF assessments on carriers and their customers. The
contribution base problem stems in large part from the wording of the Act itself.
gardless of the number of lines served, the loss of lines by the incumbent will increase per line
support, all other things being equal.
22 Universal Service Administrative Company, Federal Universal Service Support Mechanisms
Fund Size Projections for the First Quarter 2007 (Nov. 2, 2006), App. HC01. Ironically, if AT&Ts
support in Mississippi was determined under the rural support mechanism, its support for 2007
would fall from $101.2 million to $24.7 million. See, National Exchange Carrier Association,
Submission of 2005 USF Study Results (Sept. 29, 2006), App. E. Because of the equal support
rule, the support paid to competitive ETCs would fall as well.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00032

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

29
Section 254(b)(4) states that: All providers of telecommunications services should
make an equitable and nondiscriminatory contribution to the preservation and advancement of universal service. However, Section 254(d) states: Every telecommunications carrier that provides interstate telecommunications services shall
contribute on an equitable and non-discriminatory basis, to the specific, predictable,
and sufficient mechanisms established by the Commission to preserve and advance
universal service. In other words, even though the principle set forth in the Act
is that all telecommunications providers should contribute to the fund, and even
though the Fund benefits all areas of the country, Section 254(d) limits the obligation to support the Fund to a subset of telecommunications carriersproviders of
interstate telecommunications services.23
In 1997 the FCC decided to base the funding for the high-cost and low-income
support mechanisms on each carriers interstate and international revenue, while
the funding for schools and libraries and rural health support mechanisms were
supported by assessments on all revenues, interstate and intrastate. The use of
intrastate revenues for USF assessment purposes was struck down by the Fifth Circuit Court of Appeals in 1999.24 Since that time the contribution base for the USF
has been limited to only interstate and international revenues. As the USF has
grown, and as the interstate revenue base has leveled off, the assessment rate has
increased rapidly.
So long as interstate revenues grew at a reasonable rate, the ultimate impact of
fund growth on the USF assessment rate and customers bills was fairly moderate.
However, beginning in 2000 interstate revenue growth began to flatten out, and
during 2002 started to decline. The result has been a steep escalation in the USF
assessment rate, from 5.7 percent in the fourth quarter of 2000 to 9.7 percent in
the first quarter of 2007.25 Based on the latest projections from USAC, the assessment factor for the second quarter of 2007 is likely to exceed 11 percent.
There are several alternatives available in order to broaden the USF contribution
base. One alternative would be to retain the current system, but remove restrictions
in current rules which artificially depress the existing interstate revenue contribution base. One such restriction is the so-called safe harbors which limit the contribution responsibility of certain classes of carriers, such as wireless carriers and
Voice over Internet Protocol (VoIP) carriers. Another restriction limits the contributions from broadband providers, one of the fastest growing areas of telecommunications. Currently, providers of broadband are exempt from paying to support the
USF.26 If the Commission includes broadband in the list of USF supported services,
it is obvious that broadband providers should also contribute to the Fund.
A second alternative would be to grant the FCC the authority to base contributions to the Fund on total telecommunications revenues. Shown on Attachment 6 is
a comparison of changes in the Universal Service Fund, the interstate revenue base,
and total telecommunications revenues from 1997 to 2007.27 As you can see, total
telecommunications revenues currently amount to approximately $230 billion and
would provide an adequate funding base for the USF. In fact, if total telecommunications revenues had been used as the funding base from the start, we would not
be discussing this issue today. The growth in the Fund could have been accommodated while keeping the assessment rate around 3 percent.
Use of total revenues would also eliminate disputes about whether revenues are
intrastate or interstate, and would equitably spread the obligation to support Universal Service to all providers and to all customers based on their use of the network. However, basing Federal Universal Service on total revenues would require
a statutory change to clarify that the FCC has the authority to base contributions
on all revenues, intrastate as well as interstate. In addition, a total revenues base
could be susceptible to erosion in the future as more and more traffic, including
voice traffic, migrates to the Internet and is classified as information services, cur23 As a practical matter, virtually all telecommunications carriers provide some sort of interstate service.
24 Texas Office of Public Utility Counsel v. FCC, 183 F.3d 393 (5th Cir. 1999) at 448.
25 These increases have been flowed through to most customers by means of line items. Beginning in the second quarter of 2003, carriers can no longer mark up these assessments, but can
only flow through the assessment rate approved by the Commission.
26 Digital subscriber line service (DSL) providers previously paid into the Fund, but were exempted by FCC action in 2006.
27 On Attachment 6 USF Funding and the Interstate Revenue Base are taken from USAC reports. The Total Revenue Base is taken from the FCCs Trends in Telephone Service reports.
The funding base for 1997 is estimated. Beginning in the second quarter of 2003, the USF funding base has been based on carriers projected revenue collections.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00033

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

30
rently exempt from USF assessment.28 Finally, in order to prevent any uncertainty
concerning state authority, any statutory change to allow assessment of total telecommunications revenues for the Federal Fund should specify that states have the
reciprocal right to use total revenues as the basis for assessments for state Universal Service programs.
A third alternative would be to base assessments on connections to the public
switched telephone network, or on assigned telephone numbers. The FCC has considered several such proposals over the past few years. While these connectionbased or numbers-based proposals do enlarge the base of the USF, and minimize
problems with classification of services or revenues as information services, they do
have several flaws: (1) each proposal radically shifts the funding of the USF among
industry groups; (2) each proposal appears to exempt pure providers of interstate
long distance from making any contribution to the Fund in contravention of the
plain wording of Section 254(d); (3) each proposal requires capacity-based connection
equivalents for high-capacity customers; and (4) each proposal shifts responsibility
for payment of USF charges from high-use to low-use customers.
A final alternative, which my office has proposed to the FCC, would be a hybrid
of the proposals described above. For example, the Commission could continue to
base 50 percent of the Universal Service assessment on interstate revenues, and assess the remaining 50 percent on end-user connections to the public switched network. Such a hybrid would not require a statutory change and would ensure that
all providers of interstate services, even those that did not provide end-use connections, would continue to contribute to support universal service. In addition, this 50/
50 hybrid approach would mitigate impacts on low-usage customers, and result in
contributions from various industry sectors that are very close to those produced by
use of total telecommunications revenues.
In this regard, I should note that Section 101(a) of the Universal Service for Americans Act is particularly helpful. Section 101(a) empowers the Commission to assess
for Universal Service based upon interstate revenues, intrastate revenues, connections, numbers, capacity or any combination of these methods. In short, Section
101(a) provides the Commission with a full set of tools to address different contribution circumstances that may arise as the telecommunications marketplace evolves.29
Moreover, Section 101(a) also provides reciprocal flexibility for state commissions in
assessing providers to support state Universal Service funds.
In finding a solution to the contribution base problem, I agree with Senator Stevens of Alaska who has previously said: All companies that use the network, in my
judgment, should contribute to universal service, regardless of the type of service
they provide. 30 I believe we must expand contribution responsibility to encompass
all revenues and all services that connect to the telecommunications network. Since
all benefit, all should contribute.
V. Conclusion
In order to be stable and sustainable in the long-term, the USF must be configured like a pyramid: it must have a broad and stable base of contributions at the
bottom, and a narrow but sufficient distribution of support at the top. The current
Universal Service Fund requires work on both ends of this structure. Issues related
to the contribution base must be resolved. Since all benefit, all should contribute.
In addition, the limited resources of the Fund must be properly distributed and targeted to carry out the purposes of the Act. In order to continue the public policy
success of the Universal Service Fund, we must support access, not excess.

28 It should be noted that the FCC already has the discretionary power under 254(d) to require
contributions from any other provider of interstate telecommunications if the public interest so
requires.
29 For this same reason, I oppose Section 206 of the Universal Service for Americans Act,
which prohibits the use of primary lines in distributing support. As discussed above, the major
problem confronting the Fund currently is on the distribution side. Congress should broaden,
not limit, the tools available to the Commission in addressing the problems of adapting the USF
to competition.
30 TR Daily, March 26, 2003.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00034

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

31
ATTACHMENT 1
Federal Universal Service Support
[Ranked by Support in Each State]
[2005 Disbursements in Millions]
High cost
support

Low
income
support

98.9
230.0
51.8
209.3
120.2
178.7
111.7
91.5
130.2
141.0
120.3
111.2
65.5
133.8
63.5
109.3
113.4
80.2
74.6
94.4
37.8
85.2
54.7
83.6
87.3
76.3
90.3
53.6
79.3
77.8
58.5
68.5
76.7
56.6
66.3
62.7
55.9
55.1
56.6
1.3
28.8
35.2
3.6
29.6
23.6
29.5
2.2
22.6
19.2
4.3
0.0
0.0
8.7
2.3
0.7
0.3

304.7
72.3
52.5
3.6
32.4
3.1
8.3
17.8
8.8
2.4
7.4
2.4
19.2
13.3
9.3
3.2
6.0
14.5
20.3
19.8
35.0
5.4
6.1
7.5
2.3
2.9
6.2
11.4
3.5
7.3
10.7
7.3
2.6
5.7
0.7
3.8
2.4
3.9
1.4
14.5
8.8
2.8
14.3
4.1
2.9
0.7
5.3
0.2
0.4
0.5
0.9
4.6
0.6
0.1
0.1
0.3

3,824.2

808.5

State

Rural
health
support

Schools &
libraries
support

Total
support

0.5
0.1
0.0
0.1
0.1
0.3
0.1
0.1
1.0
0.1
14.9
0.0
0.1
0.0
0.2
0.0
0.8
0.2
0.7
0.1
0.0
0.1
0.1
0.7
0.3
0.0
0.2
0.7
0.1
0.5
0.3
0.0
0.5
0.1
0.1
0.5
0.7
0.2
0.1
0.0
0.1
0.0
0.0
0.0
0.4
0.3
0.0
0.1
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0

220.8
274.2
298.3
29.4
44.0
10.6
50.1
53.4
21.0
15.7
15.9
41.5
67.1
3.0
73.4
28.0
19.9
37.0
36.0
16.7
57.4
36.3
59.5
26.5
25.2
27.6
10.1
34.7
11.3
5.4
17.8
11.4
3.8
12.5
7.7
3.0
6.3
2.8
0.7
39.4
9.1
1.2
21.0
3.2
7.5
1.8
19.3
3.9
3.1
12.7
10.8
6.9
1.7
2.4
1.4
0.4

624.9
576.6
402.6
242.4
196.7
192.7
170.2
162.8
161.0
159.2
158.5
155.1
151.9
150.1
146.4
140.5
140.1
131.9
131.6
131.0
130.2
127.0
120.4
118.3
115.1
106.8
106.8
100.4
94.2
91.0
87.3
87.2
83.6
74.9
74.8
70.0
65.3
62.0
58.8
55.2
46.8
39.2
38.9
36.9
34.4
32.3
26.8
26.8
22.7
17.5
11.7
11.5
11.0
4.8
2.2
1.0

25.5

1,861.8

6,520.0

$ Millions

1 California
2 Texas
3 New York
4 Mississippi
5 Oklahoma
6 Kansas
7 Georgia
8 Florida
9 Wisconsin
10 Arkansas
11 Alaska
12 Louisiana
13 Pennsylvania
14 Puerto Rico
15 Illinois
16 Alabama
17 Minnesota
18 North Carolina
19 Arizona
20 Washington
21 Ohio
22 Missouri
23 Tennessee
24 Kentucky
25 Virginia
26 South Carolina
27 Iowa
28 Michigan
29 Colorado
30 South Dakota
31 New Mexico
32 Oregon
33 Montana
34 Indiana
35 West Virginia
36 North Dakota
37 Nebraska
38 Idaho
39 Wyoming
40 New Jersey
41 Maine
42 Vermont
43 Massachusetts
44 Nevada
45 Utah
46 Hawaii
47 Connecticut
48 Virgin Islands
49 Guam
50 Maryland
51 D.C.
52 Rhode Island
53 New Hampshire
54 American Samoa
55 N. Mariana Is.
56 Delaware
Total

Note: Numbers may not add due to rounding. Annual support amounts less than 50,000 show as 0 due to
rounding. Support amounts shown are actual amounts disbursed. Amounts assessed and collected may be higher.
Source: USAC 2005 Annual Report NECA 2005 Annual USF Filing.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00035

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

32
ATTACHMENT 2
Federal Universal Service Support
[Ranked by Support in Each State]
[2005 Disbursements in Millions]
High
cost
support

Low
income
support

State

Rural
health
support

Schools
&
libraries
support

Total
support

$ Millions

1 American Samoa
2 Virgin Islands
3 Alaska
4 Guam
5 South Dakota
6 Wyoming
7 North Dakota
8 Mississippi
9 Montana
10 Kansas
11 Puerto Rico
12 Arkansas
13 Oklahoma
14 Vermont
15 New Mexico
16 N. Mariana Is.
17 Idaho
18 Nebraska
19 West Virginia
20 Iowa
21 Louisiana
22 Alabama
23 Kentucky
24 Maine
25 Wisconsin
26 Minnesota
27 Arizona
28 Texas
29 South Carolina
30 Hawaii
31 Oregon
32 Missouri
33 Tennessee
34 Washington
35 Georgia
36 Colorado
37 New York
38 Utah
39 California
40 Nevada
41 North Carolina
42 Virginia
43 Rhode Island
44 Indiana
45 Pennsylvania
46 Ohio
47 Illinois
48 Michigan
49 Florida
50 D.C.
51 New Hampshire
52 Connecticut
53 Massachusetts
54 New Jersey
55 Maryland
56 Delaware
Total

Total lines

Monthly
support
per line

2.3
22.6
120.3
19.2
77.8
56.6
62.7
209.3
76.7
178.7
133.8
141.0
120.2
35.2
58.5
0.7
55.1
55.9
66.3
90.3
111.2
109.3
83.6
28.8
130.2
113.4
74.6
230.0
76.3
29.5
68.5
85.2
54.7
94.4
111.7
79.3
51.8
23.6
98.9
29.6
80.2
87.3
0.0
56.6
65.5
37.8
63.5
53.6
91.5
0.0
8.7
2.2
3.6
1.3
4.3
0.3

0.1
0.2
7.4
0.4
7.3
1.4
3.8
3.6
2.6
3.1
13.3
2.4
32.4
2.8
10.7
0.1
3.9
2.4
0.7
6.2
2.4
3.2
7.5
8.8
8.8
6.0
20.3
72.3
2.9
0.7
7.3
5.4
6.1
19.8
8.3
3.5
52.5
2.9
304.7
4.1
14.5
2.3
4.6
5.7
19.2
35.0
9.3
11.4
17.8
0.9
0.6
5.3
14.3
14.5
0.5
0.3

0.0
0.1
14.9
0.0
0.5
0.1
0.5
0.1
0.5
0.3
0.0
0.1
0.1
0.0
0.3
0.0
0.2
0.7
0.1
0.2
0.0
0.0
0.7
0.1
1.0
0.8
0.7
0.1
0.0
0.3
0.0
0.1
0.1
0.1
0.1
0.1
0.0
0.4
0.5
0.0
0.2
0.3
0.0
0.1
0.1
0.0
0.2
0.7
0.1
0.0
0.0
0.0
0.0
0.0
0.0
0.0

2.4
3.9
15.9
3.1
5.4
0.7
3.0
29.4
3.8
10.6
3.0
15.7
44.0
1.2
17.8
1.4
2.8
6.3
7.7
10.1
41.5
28.0
26.5
9.1
21.0
19.9
36.0
274.2
27.6
1.8
11.4
36.3
59.5
16.7
50.1
11.3
298.3
7.5
220.8
3.2
37.0
25.2
6.9
12.5
67.1
57.4
73.4
34.7
53.4
10.8
1.7
19.3
21.0
39.4
12.7
0.4

4.8
26.8
158.5
22.7
91.0
58.8
70.0
242.4
83.6
192.7
150.1
159.2
196.7
39.2
87.3
2.2
62.0
65.3
74.8
106.8
155.1
140.5
118.3
46.8
161.0
140.1
131.6
576.6
106.8
32.3
87.2
127.0
120.4
131.0
170.2
94.2
402.6
34.4
624.9
36.9
131.9
115.1
11.5
74.9
151.9
130.2
146.4
100.4
162.8
11.7
11.0
26.8
38.9
55.2
17.5
1.0

10,872
69,425
414,396
67,059
348,183
289,052
347,899
1,328,966
506,462
1,380,168
1,180,127
1,371,860
1,732,719
407,202
940,723
24,480
714,999
815,003
980,333
1,540,622
2,268,720
2,275,897
2,003,264
808,894
3,089,638
2,703,043
2,577,209
11,590,562
2,174,893
665,486
1,933,674
3,247,315
3,085,923
3,419,234
4,611,880
2,606,818
11,284,257
1,056,543
21,285,036
1,267,684
4,596,547
4,290,319
491,107
3,492,042
7,345,084
6,372,077
7,323,440
5,688,091
10,356,878
791,292
754,305
2,135,021
3,779,199
5,983,090
3,606,266
546,439

36.79
32.17
31.87
28.21
21.78
16.95
16.77
15.20
13.76
11.64
10.60
9.67
9.46
8.02
7.73
7.49
7.23
6.68
6.36
5.78
5.70
5.14
4.92
4.82
4.34
4.32
4.26
4.15
4.09
4.04
3.76
3.26
3.25
3.19
3.08
3.01
2.97
2.71
2.45
2.43
2.39
2.24
1.95
1.79
1.72
1.70
1.67
1.47
1.31
1.23
1.22
1.05
0.86
0.77
0.40
0.15

3,824.2

808.5

25.5

1,861.8

6,520.0

165,977,717

3.27

Note: Numbers may not add due to rounding. Annual support amounts less than $50,000 show as $0 due to
rounding.
Support amounts shown are actual amounts disbursed. Amounts assessed and collected may be higher.
Source: USAC 2005 Annual Report NECA 2005 Annual USF Filing.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00036

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

33
ATTACHMENT 3
Net Universal Service Support Payments by State: 2005
[Annual Payments and Contributions in Thousands]
[Sorted by Net Support Received]
Payments from USF to service providers*
State or
jurisdiction

Schools &
libraries

Rural
health
care

Total

Estimated
contributions**

Estimated
net
dollar
flow***

Florida
New Jersey
Maryland
Pennsylvania
Illinois
Massachusetts
Ohio

$91,450
1,332
4,327
65,504
63,506
3,634
37,754

$17,761
14,530
502
19,156
9,291
14,270
35,022

$53,437
39,404
12,644
67,149
73,442
20,954
57,444

$107
0
0
75
196
0
45

$162,755
55,266
17,473
151,884
146,435
38,858
130,265

$474,550
246,120
147,285
276,859
267,388
157,471
224,776

$311,795
190,854
129,812
124,975
120,953
118,613
94,511

California
Michigan
Virginia
Connecticut
North Carolina

98,866
53,575
87,312
2,249
80,179

304,668
11,425
2,257
5,315
14,504

220,789
34,722
25,263
19,307
36,946

456
694
299
0
149

624,779
100,416
115,131
26,871
131,778

716,580
187,795
193,412
100,797
200,447

91,801
87,379
78,281
73,926
68,669

56632
111,693
29,639
79,277
259

5,716
8,282
4,075
3,514
277

12,516
50,126
3,166
11,256
377

112
114
36
120
0

74,976
170,215
36,916
94,167
913

122,711
212,680
68,888
121,551
24,842

47,735
42,465
31,972
27,384
23,929

8,732

632

1,736

11,102

34,363

23,261

0
23,579
94,387
44

893
2,927
19,823
4,622

10,840
7,542
16,679
6,925

0
363
64
0

11,733
34,411
130,953
11,591

31,241
49,090
145,534
22,577

19,508
14,679
14,581
10,986

Indiana
Georgia
Nevada
Colorado
Delaware
New Hampshire
Dist. of Columbia
Utah
Washington
Rhode Island
Tennessee
New York
Missouri
Northern Mariana Is.
Hawaii

54,684
51,833
85,146

6,141
52,544
5,396

59,517
298,250
36,291

61
6
118

120,403
402,633
126,951

125,508
406,561
126,036

5,105
3,928
915

668
29,525

85
694

1,364
1,812

0
277

2,117
32,308

1,056
28,039

1,061
4,269

American
Samoa
Oregon
Arizona
South Carolina
Maine

2,318
68,469
74,550
76,322
28812

60
7,307
20,310
2,869
8,795

2,421
11,394
36,008
27,579
9,099

0
22
675
41
49

4,799
87,192
131,543
106,811
46,755

184
82,192
125,949
95,834
29,995

4,615
5,000
5,594
10,977
16,760

Guam
Virgin Islands
Vermont
Nebraska
Idaho

19,165
22,618
35,244
55,890
55,055

421
158
2,842
2,406
3,923

3,093
3,976
1,236
6,254
2,797

0
102
20
746
153

22,679
26,854
39,342
65,296
61,928

3,402
6,739
16,024
37,675
32,363

19,277
20,115
23,318
27,621
29,565

West Virginia
Minnesota
Kentucky
New Mexico
Wyoming

66,318
113,352
83,600
58,511
56,598

710
5,993
7,537
10,655
1,395

7,658
19,911
26,481
17,819
684

91
845
720
293
100

74,777
140,101
118,338
87,278
58,777

42,624
106,743
80,627
45,014
14,719

32,153
33,358
37,711
42,264
44,058

Alabama
Iowa
Wisconsin
North Dakota
Montana
Louisiana

109,343
90,336
130,225
62,718
76,731
111,241

3,224
6,198
8,829
3,804
2,631
2,414

28,023
10,042
21,021
2,956
3,807
41,487

19
186
940
503
542
5

140,609
106,762
161,015
69,981
83,711
155,147

95,271
60,490
111,194
14,669
23,456
90,833

45,338
46,272
49,821
55,312
60,255
64,314

South Dakota
Puerto Rico
Arkansas
Oklahoma
Kansas

77,788
133,786
140,997
120,188
178,684

7,280
13,286
2,369
32,358
3,149

5,434
2,966
15,662
44,003
10,545

469
0
120
129
290

90,971
150,038
159,148
196,678
192,668

15,846
52,930
58,606
74,099
58,672

75,125
97,108
100,542
122,579
133,996

Alaska
Texas
Mississippi

120,274
230,017
209,251

7,374
72,330
3,619

15,909
274,218
29,364

14,949
132
133

158,506
576,697
242,367

22,070
434,538
58,511

136,436
142,159
183,856

$3,824,187

$808,568

$1,861,745

$25,568

$6,520,068

$6,605,426

$85,358

Total

VerDate Nov 24 2008

Lowincome
support

High-cost
support

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00037

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

34
ATTACHMENT 4
Monthly Net USF Payments per Loop 2005
[Sorted by Net Payments Per Loop]
State or jurisdiction

Monthly
net payments
per loop

Delaware
Maryland
Connecticut
New Jersey
Massachusetts
New Hampshire
Florida
Nevada

530,802
3,483,388
1,997,944
5,577,359
3,529,151
719,375
9,875,661
1,248,633

$23,929,000
129,812,000
73,926,000
190,854,000
118,613,000
23,261,000
311,795,000
31,972,000

$3.76
3.11
3.08
2.85
2.80
2.69
2.63
2.13

Rhode Island
Dist. of Columbia
Virginia
Pennsylvania
Illinois

431,042
766,942
4,097,788
7,034,040
6,944,463

10,986,000
19,508,000
78,281,000
124,975,000
120,953,000

2.12
2.12
1.59
1.48
1.45

Michigan
Ohio
North Carolina
Indiana
Utah

5,105,300
5,887,158
4,362,919
3,317,961
1,022,713

87,379,000
94,511,000
68,669,000
47,735,000
14,679,000

1.43
1.34
1.31
1.20
1.20

Colorado
Georgia
Washington
California
Tennessee

2,474,508
4,416,698
3,259,380
20,610,893
2,987,705

27,384,000
42,465,000
14,581,000
91,801,000
5,105,000

0.92
0.80
0.37
0.37
0.14

New York
Missouri
Arizona
Oregon
South Carolina

10,230,291
3,081,156
2,419,556
1,855,141
2,073,761

3,928,000
915,000
5,594,000
5,000,000
10,977,000

0.03
0.02
0.19
0.22
0.44

Hawaii
Texas
Minnesota
Wisconsin
Kentucky

632,638
10,945,498
2,565,929
2,877,855
1,904,145

4,269,000
142,159,000
33,358,000
49,821,000
37,711,000

0.56
1.08
1.08
1.44
1.65

2,196,302
767,662
1,468,226
2,002,682
953,275

45,338,000
16,760,000
46,272,000
64,314,000
32,153,000

1.72
1.82
2.63
2.68
2.81

764,517
694,630
909,041
22,770
397,603

27,621,000
29,565,000
42,264,000
1,061,000
23,318,000

3.01
3.55
3.87
3.88
4.89

1,635,403
1,313,238
1,158,243
1,284,666
480,860
1,250,753

122,579,000
100,542,000
97,108,000
133,996,000
60,255,000
183,856,000

6.25
6.38
6.99
8.69
10.44
12.25

Wyoming
North Dakota
South Dakota
Virgin Islands
Guam

273,429
332,667
333,770
68,956
65,044

44,058,000
55,312,000
75,125,000
20,115,000
19,277,000

13.43
13.86
18.76
24.31
24.70

Alaska
American Samoa

389,001
10,956

136,436,000
4,615,000

29.23
35.10

151,029,353

$85,358,000

Alabama
Maine
Iowa
Louisiana
West Virginia
Nebraska
Idaho
New Mexico
Northern Mariana Is.
Vermont
Oklahoma
Arkansas
Puerto Rico
Kansas
Montana
Mississippi

Total

VerDate Nov 24 2008

Net USF
payments

USF Loops

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00038

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

35
ATTACHMENT 5

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00039

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

301gregg2.eps

301gregg3.eps

ATTACHMENT 6

36
STATEMENT OF HON. DANIEL K. INOUYE,
U.S. SENATOR FROM HAWAII

The CHAIRMAN. I thank you very much, Director Gregg. And I


thank the panel very much.
We will begin our questioning with Vice Chairman Stevens.
Senator STEVENS. Well, thank you very much, Mr. Chairman.
You havent made your opening statement. Do you want to make
the opening statement?
The CHAIRMAN. It will be made part of the record.
Senator STEVENS. Ill make mine part of the record, too, then.
[The prepared statements of Senators Inouye and Stevens follow:]
PREPARED STATEMENT

BY

HON. DANIEL K. INOUYE, U.S. SENATOR

FROM

HAWAII

Todays hearing on Universal Service returns the Committee to familiar territory.


Indeed, it was just over a decade ago that we created Section 254 of the Communications Act, which provided the Federal Communications Commission with statutory authority to create a system of explicit support to preserve and advance the
goals of universal service.
While it may be familiar territory, todays Universal Service system faces new
challenges, brought on by shifts in the way that Americans communicate and by the
steady emergence of new communications platforms.
Without question, when it comes to Universal Service reform, we face a difficult
task in balancing competing equities to promote the goals of Universal Service in
a manner that will achieve a fair result. These issues are complicated, and radical
solutions often promise more than they can deliver.
If we are to move forward in fashioning a system that is both flexible enough to
adapt to changes in the marketplace and rock-solid in its commitment to promoting
reasonably comparable communications services at reasonably comparable rates,
then all of usindustry, regulators, and Members of this Committeewill need to
roll up our sleeves, and work toward proposals that will result in meaningful
progress and a firmer footing for the stability and sufficiency of the Universal Service Fund.
I am hopeful that todays hearing, featuring two distinguished panels, will begin
this constructive discussion.
PREPARED STATEMENT

OF

HON. TED STEVENS, U.S. SENATOR

FROM

ALASKA

I would like to thank the Chairman for scheduling this hearing. Senator Inouye
has been a great leader and friend with respect to Universal Service ever since we
began working together to achieve the same phone rates in Alaska and Hawaii as
in the rest of the country.
As communications technologies advance and evolve, the mission of Universal
Service continues. The 1996 Telecommunications Act locked in certain aspects of the
Universal Service program that made sense in 1996 but that now need to be revisited. Chief among these is how the program is supported.
Last Congress, this committee worked in a bipartisan fashion to reach consensus
on how to update Universal Service contributions and ensure that all communications carriers are covered. S. 101, the USA Act, reflects the work that the Committee engaged in last year. This would result in a more rational approach with a
smaller fee on most consumer phone bills. For the elderly, who could have been disproportionately impacted, we created an exception. The bill also ensures that no
technology is excluded from being able to receive Universal Service funds. While
there seems to be consensus relative to Universal Service contributions, there are
a number of issues before the FCC relative to distributions that do not yet have consensus. I expect that we will hear about proposals for reverse auctions and other
ways to limit Universal Service spending today, and I look forward to that discussion.
Fiscal controls are important, so long as they do not undercut Universal Services
mission to deliver communications service to rural America. I will listen today to
understand how any proposal ensures that the network costs of rural carriers will
be sufficiently supported. In addition, I hope witnesses will explain what mechanisms will exist to allow carriers to make new investment to bring new and essen-

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00040

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

37
tial services to rural America. Unless these and other concerns are addressed, I do
not see how the FCC can move forward to implement some of these proposals.
I look forward to working with my colleagues to address contribution reform in
this Congress, and to better understand what options the FCC is looking at to address the distribution methodology in a way that will not disadvantage rural America.

Senator STEVENS. I congratulate the panel. You really have presented the viewpoints of your own entities and yourselves on probably the most difficult problem we face. And it does seem to me
that somehow or other we have to devise a way to have an adjustment period to get back to the point where there is just one subsidy
involved in these areas of very high cost. Has anyone got any idea
how to do that? Now, youve heard my comments before. I think reverse auctions will just do no more than bring in national concerns
thatll absolutely wipe out all the local carriers who have pioneered
these lines in the past, which I think is very unfair. But, on the
other hand, I also think that those legacy carriers have got to adjust, and theyve got to have some way to become broadband providers. Some of you have suggested that that be on the basis of
grants. So, let me ask all of you. Would you envision those grants
would come from the Fund?
Mr. BURKE. I think I was probably the one that fired that idea,
Senator. And the way I had envisioned, to this point in time, is
that, certainly, the Fund could be used as the vehicle. I felt that
a matching type of grant, using the states as the bellwether, would
be a particularly good way of doing it, because, number one, it
would give the states an incentive to be targeting extremely well;
and, number two, the states probably are the best vehicle to target.
Whether or not it was through the Fund or through a separate entity, I really have no particular opinion. But I do think, if you did
do it through the Fund, that a matching grant proposal at least has
the attraction of being focused, targeted, and, therefore, limited in
scope by that close targeting, in conjunction with the states, and
allow, therefore, the pressure on the Fund to be not undue.
Mr. GREGG. Senator Stevens, I think that the approach that was
set forth in the Universal Service for Americans Act, S. 101, section
202, which set up the broadband fund of a half a billion dollars,
was the appropriate concept that we could use for the high-cost
fund, as well. As set forth in section 202 of the Universal Service
for Americans Act, the support is limited to a single facilities-based
provider per unserved area. That is a rational way to target and
maximize the benefit of that type of program. I would conceive that
this broadband fund should be a part of the overall Universal Service Fund. And I also endorse the approach given by Commissioner
Burke; having the states have to pony up their own money, so that
its a joint State-Federal effort to bring broadband to all Americans,
I think, is a very good idea.
Obviously, policymakers deciding how much they have to spend
for a particular policy, and establishing principles to guide its distribution, is the way were used to working. One of the things thats
unusual about the high-cost fund is, theres no limit. That might
have made sense in 1996, when we knew we were going to have
to make implicit subsidies explicit, but 11 years down the road

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00041

Fmt 6633

Sfmt 6601

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

38
now, it may be time for this Congress to express its opinion as to
what upward limit there should be on the high-cost fund, as well.
Senator STEVENS. What do you think about that, Mr. Copps?
Mr. COPPS. I think if were going to go down that road of general
grants, which is not a bad idea, we have to talk, first of all, about
coordination with other programs that are out there, because we already have initiatives like Rural Utility Service grants and things
like that, so there has to be a strategy for all of this. But I think
if we just look at the Fund and how to make it credible, if we
would go to the intrastate funding that I was talking about, if we
would have broadband paying in, as well as receiving, thereby
greatly expanding the revenue base of the Fund, and given that
weve already tried to true-up wireless and VoIP and do the oversightI think we would have a fairly viable approach to Universal
Service.
Senator STEVENS. Would you change the Commissions current
position that the broadband carrier would receive support based
upon the cost of the legacy carrier?
Mr. COPPS. I think thats something we can look at. I just want
to make sure rural carriers and rural consumers have access to
what they need. So, can we maybe look at that basis? Yes. But, you
know, it really concerns me to see so many of the large companies
selling off rural exchanges. That puts the burden on any company
coming in and really put broadband out to rural consumers. The
burden falls on these rural carriers. So, yes, Im alive to having the
oversight and making sure that the distribution is disciplined, but
there has to be that ability to cover the legitimate costs of getting
advanced telecommunications to all consumers in all states.
Senator STEVENS. Mr. Landis, thanks for your comment about
my Hoosier background, but what do you think about this comment
about the concept of having the states and the Federal agencies in
a partnership, in terms of bringing on this broadband conversion?
Mr. LANDIS. Mr. Vice Chairman, I would agree with my colleagues. And I would add that I think theres strong agreement
among all of us on the Joint Board in that regard. I think in many
states the Governors are looking at broadband as an economic development tool. So, in addition to just looking at it as a communications tool, it becomes an economic development tool. I believe the
states would readily opt into the opportunity to add to and to grow
funds to develop those areas that are underserved and most difficult and most costly to reach. In Wyoming, for example, theyve
undertaken a very extensive study at a very granular level to determine not only the cost of serving all unserved areas in the state,
but to determine the least-cost mode of doing so for each unserved
area. If you coupled that, for example, with an auction which allows all intermodal competitors to bid, but set a ceiling on the cost,
which is the lowest cost, then you have a level playing field opportunity for all intermodal competitors and, at the same time, secure
build-out for the least amount of funds expended to get to those
highest-cost areas.
Senator STEVENS. Thank you very much. My times up. I dont
know that other areas have the same problem we do. Our tele-medicine, our tele-education, out tele-conferencing for disasters are all
tied to this system that currently is supported by Universal Serv-

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00042

Fmt 6633

Sfmt 6601

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

39
ice. Any disruption in that would disrupt the healthcare system,
the education system, and the overall survival system, in terms of
disasters. So, were very worried about this transition in our area,
to make sure that it doesnt dislocate the existing service as it tries
to bring on a new service.
Thank you, Mr. Chairman.
The CHAIRMAN. I thank you, sir.
Senator Dorgan?
STATEMENT OF HON. BYRON L. DORGAN,
U.S. SENATOR FROM NORTH DAKOTA

Senator DORGAN. Mr. Chairman, thank you very much.


First of all, thanks for your testimony. Senator Stevens has introduced a broadband bill. Senator Smith and I have reintroduced
our broadband bill. Both, I think, are reflective of what might have
been possible in the last Congress. I, kind of, view both as a starting point. Im much more interested in what is required than what
is possible. I think we ought to stretch what is possible to what is
required. And I want to just ask a couple of questions about that.
As I understand it, the Commission defines broadband as 200
kilobits. Is that correct?
Mr. COPPS. Thats correct.
Senator DORGAN. Thats almost unbelievable to me, frankly. You
know, you take a look at a number of foreign countries, theyre getting 20 times our speed for half our cost, because theyve developed
much more aggressive public policy, deciding that they wanted the
buildout of advanced services more universally.
I was here in 1996, with my colleagues, when we wrote this bill,
and we talked about, in this legislation, that we wanted to provide
consumers in rural and high-cost areas with access to telecommunications and information services that are reasonably comparable to
those services and rates provided in urban areas. And so, what we
said was, Individuals in rural, insular, and high-cost areas should
have access to basic and advanced services at rates that are reasonably comparable to rates charged for similar services in urban
areas. I dont understand, I would say to the two Commissioners,
why the Commission has described that provision in law as not including Universal Service support for broadband.
Mr. Copps, could you respond to
Mr. COPPS. I agree. I think we have the charge, under the Telecommunications Act, to be addressing that and bringing advanced
services to rural areas. As I said in my statement, I dont know
that we have a consensus on that particular challenge, but Im certain that we need to meet it. This is the infrastructure of our future. Im convinced that some of those rural citizens in North Dakota are just not going to get high-speed broadband, really competitive high-speed broadband, unless we develop a national strategy
and use this Fund in a more effective fashion to bring that kind
of advanced telecommunications to all of the citizens of your State,
and the other states, too.
Senator DORGAN. Commissioner Tate?
Ms. TATE. You know, I share your concerns. I think we all do.
We are very concerned about getting these services out to the rural
parts of America. I think that, really, Commissioner Landis hit the

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00043

Fmt 6633

Sfmt 6601

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

40
nail on the head, in that this is really about economic development
for our whole country. So, we look forward to working with you as
we move forward.
Senator DORGAN. But my question wasnt about my concern. I
think all of us have the concern. My question was about the way
we wrote the law. We specifically included, in the law, advanced
services. We werent sure what those advanced services were, but
we understood what we wanted Universal Service to mean; and
that is, people in rural areas would be able to get reasonably comparable services at affordable prices. And so, I think the way this
is wordedyou know, I went to a high school where I had a senior
class of nine students. We didnt have a foreign language. So, I
couldnt read it if it were a foreign language. But this is not a foreign language. As I read this, it says advanced telecommunication
and information services shall be accessible, in all regions of the
Nation, and then we talk about access to advanced services at
rates reasonably comparable. I dont understand how that can have
been misread for so long by so many, Mr.
Mr. GREGG. Senator Dorgan, 254(c) of the Telecom Act sets forth
how Universal Service, or those services that will be supported by
the Universal Service Fund, are to be determined. Universal Service is defined as an evolving level of telecommunications services.
And there are set forth a number of criteria which the Joint Board
has to evaluate in determining whether to add services to the list
of supported services. Currently, there are, I believe, 13 separate
services, which basically make up plain old telephone services, that
are supported. One of the criteria that we have to look at in deciding whether to add broadband is whether it is subscribed to by action of market forces by a substantial majority of residential customers. The last time we examined this, in 2002, only 12 percent
of residential customers actually subscribed to broadband. The verdict of the Joint Board then was, Its coming, but its not yet
there. Right now, we are almost at 50 percent. The time to move
is now, under the existing law.
Senator DORGAN. You know, the Joint Board, thats a different
subject, perhaps for another day. I know it relates to this, but the
Joint Boards been talking about primary line restriction for rural
areas, and so on. What that is, is a carve-out for a disadvantage
for rural areas. So, I have minimum high regard for some of those
recommendations, I might say.
Let me ask, my understanding is, the FCC currently says that
broadband is deployed, in an area if one customer in a Zip Code
is served. Is that correct, Commissioner Copps?
Mr. COPPS. That is correct. Its almost like saying if one person
in your town drives a Mercedes-Benz, everybody must drive a Mercedes-Benz. But thats the methodology weve used. I think the
Commission is on the verge of trying to tee up a little more sophisticated approach to this, but that involves notice and comment on
a proceeding, and were about 15 years too late to be heading down
that road.
In answer to the previous question, though, a further factor that
you need to take care of is, weve been doing our dead-level best
not me, but the Commission, generallyto exclude some of these
systems, some of these technologies, from Universal Service by re-

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00044

Fmt 6633

Sfmt 6601

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

41
classifying them as information service. So, the plain old telephone
service of the last century, yes, were supporting that, the POTS.
But we dont support the PANS, which is the Pretty Awesome New
Stuff. And I think America needs the POTS and the PANS in the
21st century.
Senator DORGAN. Well, first of all, this is horribly complicated;
I understand that. I mean, just trying to think through all of this,
and understand itits horribly complicated. But, Ive got to tell
you, I think were really tiptoeing, and have been tiptoeing for a
long while; and, in some cases, been tiptoeing on not very solid
ground. I think some serious mistakes have been made in implementation of the law. No question, we have to broaden the base.
All of us understand that. And there have been previous decisions
that should have broadened the base, that did not. We have to do
that.
But, more than that, I think both the Congress and the Commission have to decide on how broad these goals are going to be, and
how aggressive were going to be in meeting them. And I sense a
reluctance here in the panel and also with respect to the witnesses,
a reluctance to really describe what is required. Because I come
from a town of 300 people. Im just telling you, lots of places are
getting left behind and are going to be permanently left behind.
Mr. Landis, you talked about this being economic development.
Well, Ill tell you where economic development is not going to happen, certainly if the Joint Board would have had something to say
about primary line. I mean, you will not have any crack at economic development in areas where you live on the wrong side of
the digital divide.
So, Mr. Chairman, thank you for holding this hearing. I think
this is a starting point, and Im much more interested in what is
required
Senator STEVENS. Would you yield for just one
Senator DORGAN. I certainly would.
Senator STEVENS. Senator, if you pick up the phone and want to
make a reservation in one our major hotels, youre talking to someone in India. That system could be in North Dakota or in Alaska,
in many parts of rural America, but for the absence of this kind
of service. I really think it is economic development, and bringing
home some of the stuff thats gone overseas, because of the lack of
the communication network we needed to keep up.
Senator DORGAN. Well, I agree with that. All Im saying is, I
dont think any of usthe Congress, the Commissionnone of us
can afford to be timid. I mean, weve got to move. Its been 11 years
now since we passed this bill.
The CHAIRMAN. Yes.
Senator DORGAN. And, as I said, I read it, and its in English,
and I understand what is possible. I want the FCC to broaden its
capability to read this in a way that does what is possible to give
all Americans the same opportunities.
Mr. COPPS. Can I make just one comment? Because I dont want
anybody ever to think that Im the least bit reluctant about this.
I believe that this is absolutely integral to the future of our country. I believe this is the central infrastructure challenge that the
United States of America has. As you look throughout our history,

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00045

Fmt 6633

Sfmt 6601

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

42
weve had infrastructure challenges. When we first started this
country, settled the new lands, the question was, how do you get
products to markets? So, we decided, as a country, to build roads
and turnpikes and river improvements and canals and all the rest.
Then we built regional railroads. After the Civil War, were a continental power, how do we bring the country together? We committed
to the infrastructure, we built the transcontinental railroad. Even
in the Eisenhower years, we built the interstate highway system.
This generations rendezvous with destiny is to get these modern
telecommunication systems out to all of our citizens. Its economic
development, its individual opportunity, its individual fulfillment.
And I think its the future of the country.
Senator DORGAN. Mr. Chairman, my time is up, but Ithe interstate highway issue is exactly the point. They probably wouldnt
justify building an interstate highway through North Dakota, East
to West, except as it connects our Nation and as a bridge. The
same is true with respect to the digital interstate highway that we
have to build everywhere in this country.
The CHAIRMAN. Thank you very much.
Senator Pryor?
STATEMENT OF HON. MARK PRYOR,
U.S. SENATOR FROM ARKANSAS

Senator PRYOR. Thank you, Mr. Chairman.


Interesting discussion today. I want to thank the panel for being
here. When you look at the future of telecommunications, to me it
seems like its wireless and broadband. I dont think theres any
doubt about that. I mean, thats the way everything seems to be
going. And Im glad were having a broadband discussion here
today, because its essential. Its like what everybody has said in
the room. There are some challenges in figuring it out, but we have
to do it the right way to make sure that all Americanswithin reason, but all Americans have access to broadband.
Let me switch gears just for a moment, about wireless, because
I do feel like America is going wireless. Clearly, thats just the
trend that things are going, the direction things are going. For example, in Arkansas, we have 47,000 farms. Those farms support
about 287,000 jobs in the ag sector. Farmers need wireless communications. Someone told me, yesterday, theyd rather sit on their
tractor and transact businessits not easy for them to get off and
go in the farmhouse and do all thatand do that. And thats the
way it is everywhere. And everybody ought to have access to that.
Andfor example, a farmerbroadband is important to them;
wireless broadband. They can check commodity prices. They can
transact their business when theyre out there in the fields and
taking care of other business. And they ought to have that same
access that other people have, as well.
So, Ms. Tate, let me start with you, if I can, and really ask the
whole panel about the new rules for high-cost support that will
continue to support wireless. Is that what were committed to
doing, is to make sure the future is wireless and broadband?
Ms. TATE. Well, I think that youre right, that there are shortterm issues and solutions that weve tried to lay out here today.
And then, there are also longer-term solutions. And, as Senator

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00046

Fmt 6633

Sfmt 6601

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

43
Stevens says, weve got to think about the transition in between,
so that companies who have been reliant on some of these funds
have the opportunity to, as they are learning new business plans
and reengineering their business plans for all the new technology,
also have to realize whatever is going to happen with the Fund. So,
I certainly recognize the need, and we want to be supportive of all
the new innovative technological changes, and I appreciate exactly
what youre saying.
Senator PRYOR. You know, one of the challenges, being from a
rural Stateand pretty much all of us are from rural states here
today, or states that have a large percentage of rural population
one of the challenges in the traditional wirelineby the way, I
think theres always going to be a market for wireline, the traditional telephone. I dont see that going away. I dont think well go
100 percent wireless. But one of the challenges has always been,
in rural America, to string that copper wire out the miles and miles
and miles you have to, to get to a few customers; whereas, you
have a densely populated area, where you have to string it maybe
a block, and you get ten times the customers that you would get
with several miles out in the country. But I would like to hear from
the rest of the panel about your commitment to making sure that
wireless is a real option for rural America.
Mr. Copps?
Mr. COPPS. Well, I agree with that, too. I think what wireless has
already done is fantastic, and the future is boundless. That being
said, I think we are under a charge to observe some semblance of
technology neutrality at the Commission. The reality of the situation right now is that 98 percent of the people who are getting
broadband today, are getting it through DSL and cable modem. So,
our hope for the future is that wireless will play its rightful role.
I think we will see a lot of innovations. We already are. And I
think the one obligation of the Commission is to encourage that, to
provide the right kind of incentives that dont disadvantage another technology in the process, but that really open the frontiers
for these industries to develop.
Senator PRYOR. Right.
Commissioner Landis?
Mr. LANDIS. I would agree with my Federal colleagues that wireless has to play an important role in the process. The challenge is
to address it in such a way that we dont inadvertently make the
wrong decisions. In
Senator PRYOR. Tell us what
Mr. LANDIS.many cases
Senator PRYOR.tell us what you mean by that.
Mr. LANDIS. In many cases, in hundreds of areas across the country there are already multiple wireless companies present. In many
of those cases, those entries occurred based on a competitive model;
that is, they entered the market to serve a customer base that they
saw that they could do without support. In other cases, companies
have built their entry premised on Universal Service support. And
whatever we do, in terms of the solutions that are developed, we
want to make certain that we dont inadvertently advantage one
company in an environment which may be largely competitive. The
challenge, of course, is in separating those two out and determining

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00047

Fmt 6633

Sfmt 6601

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

44
where a truly competitive situation has been the motivating factor
behind entry, and those areas where entry really does require support.
Senator PRYOR. Actually, just a comment on that. I love competition. I think thats healthy. But weve all heard the terms like
cream-skimming or cherry-picking. Weve all heard those terms.
But, you know, we all know that the investment, by and large, is
going to follow the population. I mean, thats just the way it is. I
mean, because thats where the money is, thats where the preliminary investment is. So, I just think we need to make sure that the
proper amount of investment is going out to rural America to serve
those needs, as well.
Did you have a comment?
Mr. BURKE. Senator, I come from a State, too, where the C I
worry about a lot isnt necessarily competition, but, instead, is coverage. And
Senator PRYOR. Right.
Mr. BURKE.I understand
Senator PRYOR. We have some that
Mr. BURKE.your concern.
Senator PRYOR.yes.
Mr. BURKE. The only thing I will say, though, is that I believe
that the Universal Service Fund should have some limit as to
where the line would be drawn between subsidizing competition
and making sure that rural America has a reasonably comparable
service. Thats a challenge. Ill freely admit that. Because both
sides of that coin can readily be seen. I think that maybe thats
what we have to focus on. Maybe the focus ought to be, where do
we draw that line? And I think that thats a challenge.
Today, my son has no idea what a wireline is. He thinks just
electric service runs through those wires and poles
Senator PRYOR. Right.
Mr. BURKE.outside his door.
Senator PRYOR. Right.
And, Mr. Gregg? Thank you.
Mr. GREGG. Senator, your question about wireless raises some of
the issues with the problems of the high-cost fund today. We have
created a de facto $1 billion wireless infrastructure fund through
the operation of current high-cost fund rules. Nobody planned it
that way, nobody intended it, nobody is looking at it that way
today. But that is, in fact, what it is. The problem is, it is not distributed evenly. The wireless carriers are flocking, obviously, to
where the money is, like Willy Loman, in talking about robbing
banks. Theyre doing what is economically rational under the current rules.
If we want to support wireless buildout in rural areas, if this
Congress wants to support wireless buildout in rural areas, they
should say so. Like the broadband fund in S. 101, there probably
should be set up a wireless broadband fund, as well, with some
principles to guide distribution. Otherwise, were going to continue
to have it pocketed away in certain discrete areas instead of equitably distributed throughout the United States.
Senator PRYOR. Mr. Chairman, thank you.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00048

Fmt 6633

Sfmt 6601

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

45
Senator DORGAN. Mr. Chairman, might the Senator yield to me
just for a moment?
Senator PRYOR. Sure.
Senator DORGAN. I did not mention that Senator Pryor is, of
course, a part of Senator Smiths and my Universal Service bill
Senator PRYOR. Thank you.
Senator DORGAN.and has played an integral role in that.
Senator PRYOR. Im proud to be part of it.
Senator DORGAN. I neglected to mention that. I apologize.
Senator PRYOR. Thank you.
The CHAIRMAN. Thank you.
Senator Smith?
STATEMENT OF HON. GORDON H. SMITH,
U.S. SENATOR FROM OREGON

Senator SMITH. Back to the same bill. Im proud to be with my


colleagues on this bill. Enough may have already been said about
it, but it does create a $500 million account within the Fund, and
it specifically targets broadband deployment.
I think youve all answered this generally in agreement that the
Universal Fund should be used to promote broadband deployment.
But Im wondering, are there other ways we can encourage
broadband deployment to rural Americans? And Im really thinking, should the Universal Service Fund be tied to minimum
broadband speeds? Does anybody have a thought on that? Should
that be a standard?
Mr. COPPS. Well, I know it shouldnt be tied to 200 kilobits up
and down, as we currently define broadband. Youve got to find a
way to incentivize it. And youll hear a lot of talk today about the
wonderful job were doing with broadband deployment and penetration. But so much of it is at speeds that are not going to make us
competitive in the world, are not going to make rural America competitive with urban America. So, having some benchmarks like
that, I think, is an important part of an incentivizing system thats
really going to get this stuff out.
Senator SMITH. Well, you know, to Senator
Mr. COPPS. I mean, the devils in the details, but I think the concept is certainly worth looking at.
Senator SMITH.to Senator Stevens point, if were going to bring
the jobs back from India to Alaska and North DakotaId throw
in Oregon, too, and Arkansasits got to be comparable, doesnt it?
Yes.
Certainly, one of the goals of the Fund is to reform the control
and growth of that fund. And, you know, Senator Stevens has mentioned that the reverse auctions may or may not be a good idea.
I dont know. I think hes opposed to that. I have no firm position
on that, especially. But Im wondering, are there other ways we
might control the growth of the Fund, short of capping it or reverse
auctions?
Mr. GREGG. There are any number of ways you can limit the
Fund. The Fund has a number of discrete inputs that result in
what finally comes out of the Fund. Decisions that were made early
on, as I said in my opening statement, have resulted in the system
we have now. We could have gone to a system similar to that de-

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00049

Fmt 6633

Sfmt 6601

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

46
scribed by Commissioner Burke for the whole high-cost fund, where
we simply allocated a certain amount of money to each State, said,
If you match this, you get the Federal share. You then decide how
to allocate it. Your states are closer to it. You know where the
money needs to go. You know where the high-cost areas are. You
know where the unserved and underserved areas are. Go to it, subject to audit after the fact. You could have, even within the context of the current Universal Service Fund, limited receipt of Universal Service subsidies to only one facilities-based provider for
each study area in the United States. All of those would have constrained the size of the Fund, but yet provided adequate support
to do everything we want to do.
Senator SMITH. Is there a consensus that the best way to direct
is through the states, as opposed to a certain provider?
Mr. GREGG. I think the states have an obvious role. In fact, the
Tenth Circuit Court of Appeals, in the Quest I decision, said that
section 254 describes a cooperative State-Federal effort to promote
and advance Universal Service. So, states will always be involved.
Senator SMITH. Any other thoughts on that?
Mr. COPPS. Well, I would hope we would always have a constructive Federal-State relationship. I think weve kind of gotten away
from it in some of the FCC preemptive activities that have taken
place over the last few years. The genius of America is having that
balanced partnership.
You asked about specific steps to control the growth of that fund.
You know, one that several observers have talked about, and I
think I mentioned, is doing something about the identical support
system. Yes, we want to encourage all of these multiplicity of technologies, but we want to do it in a realistic fashion and allow for
the recovery of legitimate costs. But this fund is under too much
pressure to go beyond that and to be adding any monies that dont
need to be in it. So, thats one way that we could do it.
Senator SMITH. Thank you.
Mr. BURKE. Senator, I would only mention, and only add, that
I think that there are other ways besides just an auction that can
be viewed. In fact, weve heard, from several presenters, issues on
disaggregation and better targeting the areas that need help.
Theres no particular answer that seems to be an absolute given at
this point in time. And one of the reasons for the capsI just want
to make sure you dont think that the idea of a cap was an ultimate
solutionas a matter of fact, it clearly is a Band-Aid, trying to give
us the ability to work toward that point where we can come up
with a more permanent answer.
Senator SMITH. Thank you very much.
The CHAIRMAN. Senator Rockefeller?
Senator ROCKEFELLER. Thank you, Mr. Chairman.
I just want to make a point, since I came in late, and then I have
two questions within 7 minutes.
The just-departed Senator, Olympia Snowe, and I had an event
the other night, which many people should have been at, which
was about the E-Rate. And you cant possibly expect me to be in
the Commerce Committee on this subject without talking about the
E-Rate. I think its probably received more oversight than any government program ever conceived. And thats been a good thing. We

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00050

Fmt 6633

Sfmt 6601

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

47
have not paid the same level of attention to the high-cost fund.
Were facing serious challenges in trying to adapt Universal Service
to, as Commissioner Copps has said, an entirely different telecommunications environment.
I share Billy Jack Greggs concern that the growth of the Fund
is becoming increasingly financially burdensome to the consumers.
If we do not adopt policies that limit the growth of the Fund, we
will know that, as the Honorable Mr. Burke has indicated, that a
cap will be inevitable, but not at this point.
I would prefer that the FCC adopt policies that would limit the
growthand this will lead to a question to the two Commissionersto limit the Fund of the growth so that the cap is not necessary. But I knowthat was a very thoughtful statement by Mr.
Greggsupporting the cap, at least in the short term, because of
Congresss and the FCCs inability to make hard decisions, what it
amounts to. Senator Dorgan was talking about that. I think its unfortunate it got to this point.
The FCC and the Joint Board are facing short-term tough decisions to limit the growth of the Fund. We all must begin the long
process. I believe there are three main points. I believe that
broadband providers must begin to pay into the Universal Service
system if were going to have a long-term sustainable base of revenues, called a pyramid base. Two, we should demand that recipients of Universal Service Fund resources that get those things,
that they be required to transition their networks into the nextgeneration broadband network. That has not been necessarily advanced toward them, but I think its critical. It does not make
sense to continue subsidizing the deployment of networks that are
becoming obsolete. We have been told, third, for 2 years, that
broadband is the future of all communications. I agree with Senator Dorgan again. We talk, we talk, we talk, we talk. I think Ive
been on 12 bills in the Finance Committee to do with broadband,
none of which get it. They all get 75 cosponsors, and nothing ever
happens. So, weve been told for 2 years that broadband is the future of all communications. Weve got to make sure that rural
Americans fully participate in this future.
Now, my questions, to all panelists; in a previous hearing, Chairman Martin said that the FCC has the authority to broaden the
Universal Service program to include broadband, but, he said, it
didnt have the money. In Mr. Burkes statement, he states that
current law may not allow the FCC to include broadband in the
program. I would like to ask each of the other panelists, or all of
the panelists, starting with Commissioners Tate and Copps, their
thoughts on whether they believe the FCC has the authority to add
broadband to the list of supported services to the Universal Service
Fund.
Ms. TATE. Well, Senator, I think that whether we have the authority or not also has to be balanced with whether or not you utilize that authority or not, and the continued pressure that that
would put on the Fund. So
Senator ROCKEFELLER. You know, Im already off track with you.
The answer is not, Can you afford it? Do you have the authority?
Thats all Im asking.
Ms. TATE. Yes, sir, I think that we do

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00051

Fmt 6633

Sfmt 6601

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

48
Senator ROCKEFELLER. OK.
Ms. TATE.have the authority.
Senator ROCKEFELLER. Commissioner Copps?
Mr. COPPS. I think we have the authority. I think the 1996 Act
makes plain that we are supposed to consider evolving advanced
technologies. I think that translates into broadband, and can only
translate into broadband. So, yes, I think the Commission has that
authority.
Mr. BURKE. Senator, I think that the concern that I had was simply, as you read the Actand as the newest member of the Joint
Board, you tend to just go back and read the Act itself before you
go anywhere elseand the concern that I had was that, in fact,
there is, implicit in the subsequent section to the (b) section, that
there be a take rate of a majority of the residential households. Id
like to think that actually, however, thats almost a nonquestion for
us now, based on the definition we have of broadband services.
Even if you do look at that, and the take rate does have to be 50
percent to sustain a challenge, were there, or will be there so
quickly that, by the time were able to do anything with regard to
supported services, even if we were able to do it in a matter of just
a few months, were already going to have a 50 percent take. So,
hopefully thats a question we dont have to answer, anyway.
Senator ROCKEFELLER. Thank you.
Mr. GREGG. The answer to your question is yes, even though the
FCC has defined many of the broadband services as, quote, information services, they have also said that each of those services
cable modem, DSLhave a telecommunications component. As a
result, 254(c) would apply in determining whether they should be
added to the list of supported services. As I indicated in response
to Senator Dorgan a while ago, the problem with adding broadband
to the list of supported services and bringing it under the umbrella
of the Universal Service Fund is a problem with the wording of the
Act. We are going to start moving on adding broadband, but it will
take 2 years to get it finished. If you want it done faster, Congress
needs to change the wording of the Act.
Senator ROCKEFELLER. Is that on the record?
[Laughter.]
Senator ROCKEFELLER. Thank you.
Mr. Landis?
Mr. LANDIS. Yes, Senator.
Senator ROCKEFELLER. The answer is yes?
Mr. LANDIS. Yes.
Senator ROCKEFELLER. OK.
[Laughter.]
Senator ROCKEFELLER. I thought you were just recognizing that
I was sitting here.
[Laughter.]
Senator ROCKEFELLER. I wasnt sure.
Mr. LANDIS. Both.
Senator STEVENS. Ask Mr. Gregg for a draft of the amendment
he would like to
Senator ROCKEFELLER. Yes, thats a very good idea.
You got an amendment?

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00052

Fmt 6633

Sfmt 6601

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

49
Mr. GREGG. You could just change 254(c) to say it does include
broadband, as defined by the FCC.
Senator ROCKEFELLER. The Clerk has recorded that.
Mr. BURKE. Or with faster speeds than the FCC.
Mr. GREGG. Im assuming that the FCC is going to evolve that
definition. 200K may have made sense back in 1998, when they
first started recording the advance of broadband. It now probably
is something closer to 768K or 1 meg. In a few years, itll probably
be closer to 10 to 100 megs.
Senator ROCKEFELLER. Ive got to do my final question.
Commissioner Tate, last year the FCC relieved DSL providers
from paying into broadband. At the same time, the FCC required
VoIP companies to contribute to the Universal Service Fund. This
mandate is currently being challenged in court, where it could lose.
Again, last year the FCC increased the percentage of wireless consumer bills subject to USF assessment. Now, how have these decisions impacted the flow of revenues into the Universal Service
Fund? Is the FCC collecting more, or are they collecting less, revenues because of some of these decisions that they have made? Its
my understanding that wireless carriers are paying less in USF obligations than the industry did before you increased their safe harbor. What will the FCC do if the courts strike down the requirement relating to VoIP, as set forth by the FCC?
Ms. TATE. Well, Senator, I dont think that I have the numbers
with me today to absolutely answer exactly what youre asking and
what the exact figures are. I think that we tried to make those decisions, because they were the right decisions. Ithe DSL decision
was made before I got
Senator ROCKEFELLER. OK.
Commissioner Copps?
Mr. COPPS. I think the practical effect of the mistaken decision
that the Commission made to exclude broadband was really to create a shortfall which is probably somewhere in the area of $350
million to $500 million. Now, that doesnt mean the Fund is suddenly deficient, because its USAC that sets the size of the Fund.
It means it skews everything. It means you have to go out and
raise the monies from somewhere else. Different businesses, different consumers feel the impact. But theres no question in my
mind that it had an unhealthy effect. And if were really going to
go down this road of broadband that everybodys talking about, certainly we have to reverse course and make sure that its going to
be contributing.
Senator ROCKEFELLER. With the Chairmans permission, I think
thatSenator Dorgan, did you have a comment?
Senator DORGAN. No, Ill defer.
Senator ROCKEFELLER. OK.
Senator DORGAN. Thanks.
Senator ROCKEFELLER. Thank you.
The CHAIRMAN. Senator Snowe?
STATEMENT OF HON. OLYMPIA J. SNOWE,
U.S. SENATOR FROM MAINE

Senator SNOWE. Thank you, Mr. Chairman.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00053

Fmt 6633

Sfmt 6601

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

50
I want to welcome all of you here today. And obviously these are,
you know, complex issues, and some are more timely than others.
And I just want to be clear on the question that was posed by Senator Rockefeller, on the question of changes in the existing telecommunications law with respect to broadband. So, which would
come first? Would you be proceeding with any action in broadband
under the Universal Service Fund? Would you be pending any action by Congress? Or would you be taking your own action eventually on this question? Im not clear on that.
Mr. COPPS. I would be in favor of going ahead and taking actions
under authority that I think we have. Whether I can get three
votes to do that
Senator SNOWE. Right.
Mr. COPPS.at the Federal Communications Commission, I dont
know. But I think thats the road that I would like to go down, that
kind of initiative.
Senator SNOWE. Commissioner Tate?
Ms. TATE. Yes, Im trying to say this in a very thoughtful way,
but I think that, at the same time that we think that we have the
authority, and that we do believe that this is the direction the
country needs to move ahead, we have some pretty stark and dramatic rises in this Fund that we have to weigh in what we have
to do first. I think the growth of the Fund has got to be stemmed.
But, yes, I think that we should look at broadening the definition
to broadband.
Senator SNOWE. So, would that be contingent on action taken by
Congress, on that question? I mean, I think that thats obviously
an important issue, in terms of time frame here, as well, because
I agree with you, Commissioner Coppsand I think all of you probably share the same thoughtI mean, there is a timeliness question and an urgency when it comes to broadband deployment, ultimately. I mean, because we cant afford to wait, given, you know,
our standing in the world, for example. I mean, you know, we rank
19th or 20th in the world in terms of broadband deployment. In my
home state of Maine, 73 percent of households dont have access to
it. So, I think thats a major issue, in terms of time frame and what
the schedule is going to be, in the final analysis. And so, would you
be taking a vote anytime soon on this question? I mean, exactly
what
Ms. TATE. I think we
Senator SNOWE.whats the plan?
Ms. TATE.encourage the Chairman to put that before us.
Senator SNOWE. OK.
Mr. GREGG. Senator Snowe, as
Senator SNOWE. Yes?
Mr. GREGG.as I said earlier, it all depends on the FCC taking
action under current law. They have to act first, refer it to the
Joint Board. But thats a 2-year process. If you want it to move
faster
Senator SNOWE. And thats what you were
Mr. GREGG.its going to
Senator SNOWE.saying.
Mr. GREGG.be up to Congress.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00054

Fmt 6633

Sfmt 6601

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

51
Senator SNOWE. I see. So, if you want to do it sooner, then we
have to take Congressional action. I see. Because its the FCC first.
In terms of broadband deployment overall, Senator Stevens obviously has, in the telecommunications rewrite, a $500 million Fund
for unserved areas. What are your thoughts on that question? In
terms of trying to, you know, target and limit, you know, the deployment, because we want to contain the costs, is that a way to
go about it? If we use specific geographical information, as some
have recommended, as a way of trying to contain the growthif we
were to include broadband deployment?
Ms. TATE. I think youre probably going to hear from some presenters later, in the next panel, that are going to talk more about
targeted approaches, and I think thats something that we heard
at the en banc. We havent even had a chance to talk about the en
banc held last week, and we are still discussing that. We havent
discussed them as FCC Commissioners yet. But I think those are
interesting proposals. I do think its important to recognize there
are a lot of states that have incredible initiatives going on that are
using their own State tax incentives, for instance, and other incentives, to try to encourage broadband in their states. Kentucky is
one. Tennessee has a task force. So, I think that the states are
doing a lot.
Senator SNOWE. Commissioner Copps?
Mr. COPPS. I think a targeted approach is fine in the world in
which we live, but I think, in the final analysis, it comes down to,
how ambitious do we want to be in having a national broadband
strategy? And in terms of Universal Service, does that really mean
everybody? It seems to me it does. Does it mean reasonably comparable service at reasonably comparable prices? I think, yes. So,
I think the ultimate goal has to be every citizen of this country
having access to this kind of technology and service.
Senator SNOWE. Thank you.
Yes?
Mr. LANDIS. Senator, I think one of the things, which weve already addressed, and which youll also hear from the second panel,
is the need for more precise information with regard to the cost of
doing so. In reality, there are multiple ways in which we can encourage it. If the cost of buildout to unserved areas is not too
greatand dont ask me to put an exact number on itit may well
be that tax incentives at the State level would prove sufficient to
promote broadband buildout. If you look at the Wyoming experience, for example, where they have projected costs based on a
model for those areas that are currently unserved, the tenth docile
costs over $10,000 per household for buildout. And so, tax incentives are not going to prove sufficient in that situation. Clearly,
there needs to be a subsidy if we are going to move forward, and
if we have the will to actually make that a reality.
Senator SNOWE. I see.
Mr. LANDIS. But the first step, it seems to me, is getting a handle on the actual costs of doing so, at a much more granular level.
Senator SNOWE. I see.
Mr. BURKE. Senator, I think that states are really aware of their
shortcomings with regard to broadband and advanced services. I
think youre aware that Governor Douglas, in Vermont, has indi-

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00055

Fmt 6633

Sfmt 6601

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

52
cated his plan to make Vermont an E-State in its entirety by 2010.
Obviously, incentives are on the minds of states. And thats one of
the reasons why, it seemed to me, that to help both the Congress
and states target the matching-funds type of grants might make
sense. I think that it allows for a solid distribution. Because I
think, although this Fund is very laudable, the devils going to be
in the details, and the targeting is going to be extremely important.
Mr. GREGG. The market itself will ultimately make broadband
available to about 85 to 90 percent of all the households in the
United States. Its going to be that remaining 10 to 15 percent of
households where it is not economically feasible to have broadband
made available, absent some sort of explicit subsidy. Obviously,
adding it to the Universal Service Fund is a piece, tax breaks and
incentives are a piece, the RUS program of low-interest loans and
grants is a piece. Theres going to be a multiplicity of sources that
go into making broadband available to that final 10 to 15 percent.
And that is where our efforts should be focused.
Senator SNOWE. Thank you.
And finally, I would like to ask, since, were, celebrating the 10th
anniversary of the E-Rate, and one of the issues that obviously
emerged was the Antideficiency Act, and Senator Rockefeller and
I supported a permanent exemption. Commissioner Tate, whats
your response to that? Because, otherwise, I dont know what guarantees there are that we could stabilize the E-Rate, under the circumstances, if the ADA were to apply.
Ms. TATE. Yes.
Senator SNOWE. You agree with that. They should be a permanent exemption. Thats the only way to address it.
Ms. TATE. Well, Im not sure that I would go so far as to say its
the only way. But I appreciate your efforts to
Senator SNOWE. Yes.
Ms. TATE.stabilize that fund.
Senator SNOWE. Yes.
Any others? Commissioner Copps?
Mr. COPPS. I certainly support the
Senator SNOWE. Yes. OK.
Mr. COPPS.permanent exemption.
Senator SNOWE. Right. OK.
And finally, when it comes to, you know, broadband deployment
within the school systems and classrooms in America, do we have
any information with respect to E-Rate. Because I happen to believe that we should make the E-Rate adaptable to, you know, the
technologies of the future. And that includes, obviously, broadband
deployment and the bandwidths and platforms. Do we have any
current information with respect to how many classrooms in America that might have access to, you know, broadband deployment?
Mr. COPPS. I think were up in the 93 or more percent of classrooms connected, but then youve got to ask yourself, whats
the
Senator SNOWE. Right.
Mr. COPPS.speed of the connection? And are kids in rural
America going to expect a dial-up connection to the rest of the
world, or a true high-speed connection? So, I think the future of the
E-Rate is every bit as important as the past. Its got a long way

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00056

Fmt 6633

Sfmt 6601

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

53
to go. These are evolving and changing standards, and we have to
make sure that the E-Rate program accommodates those changes
and brings that level of communications to our kids.
Senator SNOWE. OK.
Thank you. Appreciate it.
Senator STEVENS. Mr. Chairman, could I ask just one question?
The CHAIRMAN. Sure.
Senator STEVENS. Its my understanding that in order to utilize
satellite delivery to consumers in isolated rural areas, youd need
a change in the law. Do you agree with that, Mr. Copps?
Mr. COPPS. Not that Im aware of. I have seen broadband delivered by satellite. Ive seen it in
Senator STEVENS. I mean, Universal Service payments, though.
Mr. COPPS. Oh. I am not aware of there being a problem. Ill be
happy to look into it and
Senator STEVENS. Thank you.
Mr. COPPS.see if there is, but I do not believe there is. I believe
it should accommodate all of those different technologies.
The CHAIRMAN. Id like to thank the panel very much.
I think its obvious to many that Im a member of the crystal-set
radio generation.
[Laughter.]
The CHAIRMAN. And, as a result, Im convinced of the dynamic
and ever-changing evolutionary character of communication. I am
one of those who worked upon the Telecommunications Act of 1996.
And, I think, 3 months later it was obsolete. We used the word
Internet twice in the whole bill. Now its part of our vocabulary.
I will be submitting questions to all of you, because time is of the
essence, and we have a whole panel waiting.
But this panel reminded me of an early time in my life when I
had a nice chat with a gentleman called Henry J. Kaiser, who was
making millions every day. And I asked him what is his secret, and
he said, Its very simple. I never use phrases like, This is an impossible task, or I never uses phrases like, This is too complicated.
Thats a cop-out. And Im glad that none of you have copped-out,
in the panel here, the Committee has not copped-out.
The Vice Chairman has suggested, and I agree with him, that we
should have a special, in-depth briefing on broadband, because, in
order to cope with this problem, wed better have a real good understanding of the potential, the limitations of broadband. And
were going to do that.
And so, with that, Id like to thank all of you for your contribution. Its been a great session for me. But Id like to submit my
questions, if I may.
Thank you very much.
Our next panel consists of the following: the Executive Vice
President of North Dakota Association of Telecommunications Cooperatives, Mr. David Crothers; the Director-Policy/Regulatory
Economist, Department of Law and External Affairs of the Embarq
Corporation, Dr. Brian K. Staihr; the Executive Vice President,
Corporate Secretary, and General Counsel of Alltel Corporation,
Mr. Richard N. Massey; the Executive Vice President, Public Affairs, Policy, and Communications of Verizon, Mr. Thomas J.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00057

Fmt 6633

Sfmt 6601

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

54
Tauke; and the Vice President of Public Policy, Midcontinent Communications of Sioux Falls, South Dakota, Mr. W. Tom Simmons.
Gentlemen, I thank you very much for your patience. Id like to
first recognize the Executive Vice President of the North Dakota
Association, Mr. David Crothers.
STATEMENT OF DAVID CROTHERS, EXECUTIVE VICE
PRESIDENT, NORTH DAKOTA ASSOCIATION OF
TELECOMMUNICATIONS COOPERATIVES

Mr. CROTHERS. Mr. Chairman, we thank you for the ability to appear before you today, sir.
The status of Universal Service is quite possibly the most important issue facing our industry today. Restructuring the Universal
Service program properly will be critical to determining whether all
Americans will have the opportunity to participate in the 21st century economy.
Our Nation, however, finds itself in a dilemma. Even though it
is acknowledged by all that Americans increasingly rely on more
sophisticated communications services and bandwidth for economic,
healthcare, and educational opportunities, some are looking to limit
the growth and mission of the program. While other countries are
making the investment to ensure ubiquitous broadband coverage
for their citizens, the United States remains a second-tier nation,
in terms of making a genuine commitment to broadband deployment. It is our position that rather than contemplating ways to cap
the Fund, or to otherwise limit the program, policymakers should,
instead, be looking for ways to enhance it and help accelerate such
deployment.
Mr. Chairman, NTCA developed a national communications policy course that would move the Nation in that direction. The plan
is forward-looking and addresses our Nations communications
needs, especially those in high-cost rural areas of our country. The
plan envisions the Universal Service program having an ongoing
mission.
To see what the Universal Service program has accomplished, I
ask that you look at North Dakota. We believe the state is a perfect
example of everything that is right with the Universal Service program. Ours is a very low-density state. Independent rural telephone companies serve over 96 percent of its geographic territory.
That wasnt always the case, however. In 1996, large out-of-state
telephone companies began selling their highest-cost exchanges. In
total, some 90 exchanges were sold in the State over a 5-year period. Locally owned, locally operated telephone companies stepped
up and bought every one of those exchanges. The result has been
greater levels of investment and greater technology for rural residents in North Dakota. Today, high-speed broadband is provided in
290 communities through a variety of technologies by independent
telephone companies.
Rural communications providers have worked hard to ensure
that rural America will not be relegated to being a communications
backwater. This approach is a stark contrast to the array of interests that wish to mold the program into something that it was
never intended to be: a mechanism to ensure competitive neutrality

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00058

Fmt 6633

Sfmt 6601

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

55
or to create government-subsidized competition where there would
otherwise have none existing.
This blind pursuit of competition for competitions sake has allowed the Universal Service program to be accessed by those who
have no real commitment to the policy of Universal Service. Even
FCC Chairman Martin and Verizon, both proponents of alternative
approaches to controlling the programs growth, have acknowledged
the cause of the growth is the CETC segment of the industry. Consequently, the Funds growth has been rapid, and, some say, politically unsustainable.
Today, there are countless plans under consideration to control
the growth of the Universal Service program. Most of them, including the auction concept proposed by Verizon, ignore the real root
of the problem. NTCAs approach makes far more sense, in our
judgment: expand the base of USF contributors, strengthen the
public-interest requirements for ETC designation, and eliminate
the identical support rule. Each of these proposals could easily be
implemented and would absolutely control the programs growth.
Just last week at an FCC forum, the subject of reverse auctions
was once again cited as the solution to many of the Universal Service Funds programs. The concept of reverse auctions is to limit
support to the lowest-cost provider. This argument is the antithesis
of Universal Service. Auctions will lead us down the road of supporting the lowest common denominator. It is truly a race to the
bottom.
Rural communications providers have a quality-of-service approach to network construction, and it has allowed the people of
rural America to enjoy a state-of-the-art infrastructure at affordable rates. Alternatively, the other technologies and services, such
as wireless voice technologies and VoIP, are built and operated at
far less stringent standards, and use the infrastructure of others.
The great misconception continues to exist that wireless handsets
are communicating directly to another wireless handset or to a
tower, or through a tower to another party. Mr. Chairman, wireless
needs wires. Universal Service support ensures that there is a
state-of-the-art underlying network upon which all these services
can rely. Reverse auctions will not ensure this.
We ask, are we willing to risk allowing the almost limitless bandwidth capacities associated with wireline to be undermined? And
what will happen, we ask, with reverse auctions, when a carrier
other than the incumbent wins the auction? Without this stream
of support, the rural incumbents, in many cases, will no longer be
functional. And we fear, when repeated winners of those auctions
replace one another over and over, and the lack of investment that
will follow. Sadly, it will be the American citizen who suffers the
consequences of these short-term fixes.
Mr. Chairman, today we are on the cusp of fully moving into a
world where data, video, and mobility are the primary objectives of
consumers. The technologies of tomorrow, though, will still be reliant on the underlying wireline voice network of today.
Mr. Chairman, thank you for allowing us to appear before you
today.
[The prepared statement of Mr. Crothers follows:]

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00059

Fmt 6633

Sfmt 6601

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

56
PREPARED STATEMENT OF DAVID CROTHERS, EXECUTIVE VICE PRESIDENT,
NORTH DAKOTA ASSOCIATION OF TELEPHONE COOPERATIVES
Mr. Chairman, you have convened us here today to consider the status and future
of our national Universal Service policy and its underlying support mechanism. The
discussion surrounding this venerable policy is nothing new and indeed has persisted, and evolved, much as the program itself has, and should. While this conversation has at times been exhausting, and at others outright exasperating, the
Nations small and rural community based communications providers welcome and
embrace it nonetheless. We do so because such dialogue only serves to strengthen
and improve this long-standing national policya policy that plays a critical role
in maintaining and expanding the communications infrastructure that is so necessary to our national and economic security. So thank you Mr. Chairman for your
ongoing efforts to ensure the goal of Universal Service remains the solid cornerstone
of our national communications policy that it has always been.
Do we still need this program? The answer to that question is an emphatic yes!
More and more Americans rely on communications every day to meet their commerce, security and entertainment needs. The bar for the 21st century communications has been raised. More bandwidth must be deployed in our networks so all
American households, urban and rural alike, can benefit from education, healthcare,
and economic opportunities that are dependent upon a robust communications platform.
Other countries of the world understand the need to make a financial commitment now to ensure adequate bandwidth in their communications networks. This
will provide their citizens with opportunities for economic growth and global participation. Rather than working on ways to cap Universal Service Funds, particularly
to wireline network providers that have deployed critical backbone infrastructure,
the Congress should be looking for ways to expand the fund, thereby encouraging
an accelerated deployment of broadband facilities throughout America.
Some question the continued need for universal service. To these doubters, I invite you to visit my state of North Dakota and see the incredible accomplishments
of this program for yourself. I can, without question, assure this committee that the
Universal Service Fund is more necessary today than ever before.
It is important when discussing Universal Service to approach it from the proper
perspective. Detractors and supporters alike cannot deny that the Universal Service
system is a shining example of successful national policy. This program is largely
responsible for the extremely high communications connectivity our Nation enjoys
today. It is due to Universal Service support that virtually any American that wishes to have voice connectivity is able to. Likewise it is largely due to this program
that such connectivity is uniform in price and scope regardless of where you live.
For more than a decade now our industry has been exposed to an operating environment marked by competition and deregulation. These concepts are in many ways
in direct conflict with the policy of universal service. Universal Service of course is
about developing the appropriate policy environment to ensure all Americans have
access to communications services of an equitable price and scope. The very nature
of the Universal Service concept does not allow for the let the chips fall where they
may theory associated with competition and deregulation.
The rural segment of the industry has always understood the reality that the policies of competition and deregulation will be ineffective if simply broad brushed
across all spectrums of the marketplace. Yet, when confronted with the policies, we
have simultaneously embraced and/or tackled them with vigor. This response is in
stark contrast to the array of entities from the private and public sectors alike that
continue trying to mold the Universal Service program into something it was never
intended to bea mechanism for ensuring competitive neutrality. Herein lays the
debate about where this program stands today, and where it should go in the future.
Unfortunately, while the Congressional intent of the Telecommunications Act of
1996 that led to the emergence of these conflicting policies was quite clear the manner in which it has been interpreted is quite another story. Competitors, state regulators, the Federal Communications Commission (FCC), and yes even some of your
Congressional colleagues have upset the delicate dichotomy that was to have existed
between the distinct concepts. The result is a disastrous situation where, under the
guise of establishing an environment of competitive neutrality, the program is being
accessed by many that have no real commitment to the policy of universal service.
Consequently, its growth has been rapid and is currently at a politically
unsustainable rate which is the root of why we are here today.
Mr. Chairman, so often throughout the course of this debate, people have directed
the industry, and particularly small rural carriers, to think outside the box in our
search for solutions to the fix we find ourselves in today. The comment might be

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00060

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

57
amusing were it not so completely oblivious to our way of thinking and operating
each and every day of our existence. If I do nothing else here this morning, it is
my overarching desire to ensure that everyone participating and listening to this
discussion ultimately leaves with the recognition and understanding that rural carriers do and always will think outside the box. Truly, they have no other choice.
What segment of the industry was the first to have completely converted to digital
switched systems? What segment of the industry was a pioneer in providing wireless options to their hardest to reach customers? From what segment of the industry
did the first company to deploy an all fiber system come? What segment of the industry was the first to offer distance learning and tele-health applications? What
segment of the industry was an early leader in providing cable-based video, then
satellite video, and now IP video to their markets? What segment of the industry
quickly moved into Internet service provision in the early stages of the Internets
public evolution? And what segment of the industry continues to lead in the deployment of high-speed broadband capable infrastructure?
Mr. Chairman in every instance the answer to those questions isthe small rural
segment of the industry. Many might be asking why these carriers care or have this
unique perspective and approach to their mission. The answer to that question is
relatively simple. Because these systems are owned and operated by the members
of the community in the case of cooperatives, or by members from the community
in the case of commercial systems. Clearly as a result they are entrepreneurs. Clearly they are continually thinking outside the box.
But, does thinking outside the box mean we should automatically discount the obvious? Frankly, it is astounding to us at how great the zeal of some is to do just
that. Today there are countless plans under development and already on the table
directed at how to control the growth of the Universal Service program. They are
Byzantine in their detail and approach to eventually get to the end-point desired.
Even worse, such plans also completely ignore the most obvious, basic, and easy
to implement responses. Expanding the systems assessment basestrengthening
the requirements for receiving eligible telecommunications carrier (ETC) status
eliminating the identical support rule which provides competitors with inflated supportall concepts that could easily be implemented and that we know for certain
would produce the desired result.
Evidently not in the minds of many as was evidenced just last week at an FCC
forum as well as during the course of the NARUC meeting. A great many voices
continue to sing the praises of the reverse auction concept. This approach seeks to
limit support to the lowest cost and/or most efficient technology. This argument is
the antithesis of the goal of Universal Service which I mentioned is to ensure ALL
Americans have access to communications services that are comparable in price and
scope.
Auctions would presumably lead us down the road of supporting the lowest common denominator. Again, the exact opposite of what Universal Service was structured to accomplish. Traditionally, rural communications system have been built
and constructed to extremely exacting standards. While the law requires that rural
Americans receive no less, the Universal Service system and other cost recovery programs, as well as private financiers demand no less. This Quality of Service (QoS)
approach to network construction and management is the formula that has allowed
our industry to build and maintain the infrastructure that is an integral part of the
premiere communications system our Nation enjoys today.
However, today, many alternative technologies and services to traditional wireline
voice service are built and operated according to far less stringent standards. For
example, it is a well accepted fact that wireless voice technologies generally do not
approach the QoS standards of wireline calls. Another example is Voice over Internet Protocol (VoIP) oriented service which is even further away from meeting the
QoS standards of wireline voice service.
There is one more critically important reason for the inferior nature of some of
these alternative technologies. They do not consist entirely of their own infrastructure. For example, with regard to wireless service, a great misconception continues
to exist among policymakers and the public alike, that wireless hand sets are communicating directly to one another or directly to a wireless tower and directly from
that tower to another party. This is simply not the case. Wireless needs wires Mr.
Chairman. Whether its the wires to complete a wireless to wireline call or a wireless to wireless call, there are wires involved at some point in the calls path. The
great majority of these wires are owned and operated by the incumbent voice providers.
Likewise with the VoIP voice services we hear so much about today, these systems rely almost entirely upon the infrastructure of others, and to this point that
infrastructure has generally been the last mile connections of wireline carriers and

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00061

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

58
the Internet system. An interesting point to make here is that due to ineffective
statutes and regulations, services such as this are allowed to utilize this infrastructure that belongs to others without paying for such use. They are using the facilities
of rural providers, and are not paying to do so. Without such compensation, the ability of network owners to continue to invest in their networks is put in jeopardy.
Without such investment we will eventually reach the point at which such facilities
will not function. Without such facilities being able to effectively operate, many applications such as wireless and VoIP services would be unable to operate.
Which bring us back to reverse auctions. Universal service support ensures the
continuum of the underlying network upon which all other services rely. Auctions
fail to ensure that such support will continue to be provided. Are we willing to risk
allowing the almost limitless bandwidth capacities associated with a wireline network to be undermined simply because policymakers choose to make an easy policy
decision with wide-ranging long-term implications rather than buckling down and
confronting the real underlying issues associated with universal service? There are
other questions with the reverse auction concept as well. How will efficiency be determined and measured? Providing support to the system with the lowest upfront
costs may appear efficient today but what about over the long-term?
What happens when a carrier other than the incumbent wins the Universal Service support? Without this stream of cost recovery, most rural incumbents would be
hard pressed to remain operationally functional. What becomes of their underlying
infrastructure that is necessary to the operations of alternative technologies? What
happens in the future when other providers consistently and repeatedly emerge that
are lower cost than the prior? Do we find ourselves stuck in a process of unending
churn of providers? Wouldnt such instability destine such providers to never being
able to secure the long-term financing that is so necessary to this capital intensive
business?
Finally, are policymakers themselves really up to the challenges that reverse auctions present. Its easy to talk about a lowest cost bidder approach saving money.
However, we think parties to such an initiative would quickly realize the fallacies
behind this concept were it ever implemented. No American, whether rural or urban
based, would be well served by reverse auctions. Indeed, I would like to submit for
the record a far more extensive paper on this subject. It was prepared at the request
of the National Telecommunications Cooperative Association by Dale Lehman who
is the Director of the Executive MBA in Information and Communication Technology at the Alaska Pacific University.
No, Mr. Chairman, as I alluded earlier, there are far better, more reasonable,
more realistic, and more workable options that will best ensure the proper application and future operation of the Universal Service system. Indeed, many such ideas
and concepts were contained in the Universal Service section of the Communications
Act of 2006 that this committee marked up late last year. That legislation was the
product of input from many policymakers and many sectors of the industry. Please
allow me to just highlight its stronger provisions:
Establishing a new definition of communications service that alleviates the arbitrage of certain carriers wiggling out from under their Universal Service responsibilities;
Expanding the base of contributors to the Universal Service Fund (USF) which
will lower the overall USF assessment for all consumers;
Providing flexibility in how the FCC assesses providers for their contributions,
which allows consideration of new technologies and services as well as modern
modes of communications;
Giving states new flexibility for their appropriate management of their state
Universal Service funds;
Codifying new minimum guidelines for receiving the eligible communications
carrier status necessary to receive Universal Service support;
Permanently prohibiting the FCC from limiting Universal Service support to a
single primary line, which ensures rural Americas small businesses remain
competitive;
Permanently exempting the program from the Anti-Deficiency Act and permanently removing the private fund from the Federal budget process which would
preclude the program from experiencing future short falls or spikes in Fund assessments;
Clarifies an entity is not exempt from contributing to the system solely on the
basis that it does not receive support from the program;

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00062

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

59
Establishing that equivalent services must live up to the geographic toll rate
averaging provisions that are in current law;
Ensuring a smooth conversion resulting from any new regulations or statutes
affecting the program by requiring the FCC to adopt transition mechanisms of
not less than 5 years for any changes in the Universal Service distribution process.
Were there areas that could have been stronger? No question, after all, the entire
bill, as in most legislative instances, was a conglomeration of compromise. Yet there
was one key area that was initially stronger that was weakened as it moved
through the mark-up process. The earliest drafts of the bill directly set the stage
for the Universal Service system to begin formally supporting the deployment of
broadband and advanced services capable infrastructure. This is a key issue.
Today we are on the cusp of fully moving into a world where data, video, and mobility are the primary objectives of consumers and voice will be secondary, or even
an afterthought. Remember my earlier discussion that pointed out how most alternative technologies are reliant upon the underlying wireline voice network. Well the
same holds true here. Regardless of whether consumers are focused on voice or some
other form of communication, they will still require the underlying infrastructure
to ensure their communication gets to its destination. The only difference is that
with regard to broadband and advanced services capable infrastructure, the costs
and subsequent need for support are even greater than they are for voice only infrastructure.
There was one other omission with regard to the legislation that would have gone
a long way in controlling the growth of the program and that was the elimination
of the identical support rule. For those of you that are unfamiliar with this issue,
the FCCs rules currently allow competitive ETCs to receive Universal Service support based on the costs of incumbent carriers. So in the case of a carrier with extremely high costs, a competitor can secure a Universal Service designation for that
market and receive the exact same dollars per consumer even if their costs are a
fraction of the incumbents. It is a terrible waste of funds and is a rule that should
have been changed yesterday.
Mr. Chairman, as a concluding thought I would just like to reiterate what many
of us already know Universal Service is not. Universal service support is neither a
subsidy nor a tax. Universal service support is an industry funded cost recovery
mechanism that offsets the higher cost to build and maintain vital communications
networks in rural, sparsely populated, and insular portions of our Nation. No Federal monies are appropriated for this purpose.
America stands at a crossroads between a narrowband and broadband world. The
choice is clear. I can assure you that I and the entire rural segment of the industry
that is associated with NTCA and the other rural communications associations are
ready to work with you to move forward aggressively with a national plan to bring
broadband to all Americans as is envisioned by so many. Thank you.
ATTACHMENT A
The Use of Reverse Auctions for Provision of Universal ServiceDale E.
Lehman, Ph.D.1
This paper reviews the theoretical and applied literature on the use of reverse
auctions (also called minimum subsidy auctions or competitive auctions) for provision of universal service. It reveals that reverse auctions are feasible, and have met
with some success, for provision of new infrastructure/services into previously
unserved areas, or for the upgrading of existing infrastructure and/or services. In
contrast, the U.S. environment is one in which there are multiple existing service
providers, using a diverse set of technologies, in most supported areas. Existing infrastructure requires (i) a transition mechanism to recover past prudent investments made to serve high-cost areas; and (ii) increases the difficulty of creating an
auction that is not biased in favor of any set of current infrastructure providers
(particularly if they utilize different technologies). Unfortunately, there is scant empirical evidence on which to determine the feasibility or desirability of reverse auc1 Dale E. Lehman is Director of the Executive MBA in Information and Communication Technology at Alaska Pacific University. He has taught at a dozen universities, and held positions
of Senior Economist at Southwestern Bell Telephone Company and Member of Technical Staff
at Bellcore. He has a B.A. in Economics from SUNY at Stony Brook, and M.A. and Ph.D. degrees in Economics from the University of Rochester. He has published widely in the area of
telecommunications economics and policy, including a number of previous papers on behalf of
NTCA.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00063

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

60
tions relative to alternative methods of providing Universal Service under these conditions.
The use of auctions to award provision of utility services can be traced back to
Demsetz (1968). Demsetz introduced the notion that franchise bidding could replace
traditional public utility regulation. Particular use for provision of universal service,
or carrier of last resort (COLR) responsibilities, was first explored by Milgrom in
his 1996 Nobel lecture in honor of William Vickrey, and was first suggested for examination by the FCC in 1995. Considerable academic and practitioner work has
been conducted on auctions since that time, especially in conjunction with the widespread use of auctions for awarding the right to use spectrum resources. In addition,
there is a lengthy literature surrounding the use of competitive bidding for awarding contracts (e.g, Defense Department procurements, public works construction,
etc.) which are a discrete form of an auction (in which a single project or set of
projects is awarded on the basis of a competitive bidding process).
The use of competitive processes has a number of general beneficial properties:
they promote incentives for cost-reducing innovation, they mitigate against informational asymmetries between funding entities and entities contracted to provide services on their behalf, auctions can be used to ration scarce resources to those that
value them the most, and they can permit market forces to play a role in the determination of the quality of services provided. Competitive contracts are not a panacea, however. Victor Goldberg (1976) points out that competitive procurement and
alternative regulatory mechanisms should be compared under realistic conditions related to the nature of the service that is being provided.
Goldberg provides the example of a university food service that might be contracted out on the basis of a competitive bid, or could be provided internally by the
university itself. The latter is meant to approximate the conditions under which a
regulated utility operates. Regulators must monitor the quality and cost of service
provision, and face a number of potential inefficiencies inherent in monopoly provision by an agent with better information than the principal. Competitive bidding reduces only some of these problems, and creates some new issues. Quality of service
must still be monitored, and there are administrative costs associated with both the
awarding and oversight of contracts.
Goldberg points out that administered contracts, traditional regulation, or any
other regulatory mechanism must balance the right of consumers to be served and
the right of providers to serve. Universal Service is a statement of the publics right
to be served (at comparable rates for comparable services, in high cost and insular
areas, and for consumers of low income), and regulators become the agent of these
consumers rights. At the same time, providers have the right to an opportunity for
a competitive return on their investments.
Goldbergs key insight is that the nature of the service itself, and not the particular way in which contracts are awarded (competitive bidding or regulated monopoly, for example), is what determines the key issues that must be dealt with.
Significant investment costs raise issues associated with the need to establish longterm contracts. Volatile operating costs (e.g., fuel costs) would raise issues of risk,
regardless of the regulatory mechanism that is adopted.
This principle is pertinent to the use of reverse auctions for provision of universal
service. Provision of Universal Service entails significant investment costs (sunk
costs to a degree that depends on the technology deployed) under conditions of continual technological progress. Services are provided to consumers for which the demand falls short of the provisioning costs.2 In the U.S. there are few unserved
areas: instead, there are multiple networks, using different technologies and with
different quality attributes, and serving different parts of rural areas. There are also
a variety of regulatory restrictions placed on existing rural service providers. The
potential use of auctions must be evaluated against a backdrop of these characteristics.
This paper will review the theoretical literature and applied evidence, and is organized according to a number of related issues that must be resolved in order to implement reverse auctions for universal service. These include:
Definition of the service to be auctioned.
Size of areas to be defined.
Number of COLRs to be subsidized.
Time period for contract awards.
2 This can either result from consumer unwillingness or inability to pay the full cost of provision, or from public policy that limits their price to be less than these costs. In either case, market provision will be insufficiently forthcoming, absent some form of support.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00064

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

61
Transition/stranded investment issues.
Bidder eligibility.
Type of bidding to be conducted (sealed or open, single or multiple round,
combinatorial, etc.)
Basis for determining winning bids.
Pricing and service flexibility accompanying awards.
Monitoring and enforcement issues.
Each topic has a number of feasible alternatives. In a comparison of reverse auctions and cost proxy model-based USF, Sorana (1998) states that it can be easily
seen that the two mechanisms cannot be ranked on purely theoretical grounds. 3
Similarly, theory alone cannot determine the desirability of reverse auctions for universal service.
I examine the theoretical guidance and empirical evidence that is available from
the applications of reverse auctions in telecommunications (and some limited relevant experiences in other industries). A recurring theme will be that the complexity of these decisions increases significantly in the presence of an existing infrastructure (rather than a green-field application), and when competing service providers use different technologies (with different cost and quality characteristics).
Service Definition
The definition of Universal Service will need to be specific in terms of service
quality, coverage, and capabilities. In particular, it will need to specify whether
equal access is to be included, appropriate service quality standards (e.g., system reliability), and what data speed is to be supported. This is one area in which auctions
may be less desirable than the current USF mechanism.
Under current rules, the delivery of services can outpace the definition of universal service: for example, higher broadband speeds may be available, even while
broadband is not included within the definition of universal service. An auction
mechanism may not permit this outcomethe carriers business case will need to
support the service delivered. If policymakers want to see faster deployment, then
they will need a specific auction for their desired rate of deployment.
Broadband is not part of todays Universal Service definition, and the FCCs definition of broadband service is relatively slow by todays standards. Many rural carriers provide broadband speeds well in excess of 256k, and often in the absence of
sufficient market demand to justify the deployment costs of these higher speeds, on
a narrow profitability criterion. The justification for providing these services rests
on their economic importance to the rural community served, and the ability to provide these services is facilitated by USF.
It is precisely because of the strong cost-reducing incentives of reverse auctions
that the service definition must be precise. This means that regulators must predict
service needs at least as far into the future as the time period that the franchise
will cover. The need for such regulatory foresight undermines some of the principal
theoretical advantages of reverse auctionsthat they potentially replace regulatory
fiat with market processes.
Coverage is another key part of service definition. It is not feasible to define Universal Service as availability to 100 percent of the population. Reduced targets, such
as 90 percent, however, do not sound like Universal Service. For many years, telephone companies have operated under state-specific requirements to provide service
to any location within X miles (usually a fairly small number) of their current network facilities. Special construction charges apply to locations that exceed X, with
the costs usually borne by the party requesting service. Given that this practice has
been built into construction plans, it seems that continuing this practice would be
least disruptive to consumers.
Size of Areas
A fundamental principle for an auction to be efficient is that the item being auctioned must be the same for all bidders (their individual valuations may differ, but
the item being auctioned must be the same if the bids are to be compared). This
means that the coverage area must be the same for all COLR bidders.
Theoretical work also suggests that there may be subtle strategic effects as geographical coverage differs across competing providers. If one provider is obligated
to serve all customers at the same price, and the other carrier can serve a subset
of customers, the COLR carrier must be reimbursed for reduced profits on the contested part of the market as well as the higher costs of serving the uncontested con3 Sorana

VerDate Nov 24 2008

14:09 Mar 21, 2012

(1998) at page 18.

Jkt 073387

PO 00000

Frm 00065

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

62
sumers [Hoernig and Valletti (2003)]. The strategic considerations go further and
can raise the subsidy substantially, and even may leave both firms with higher
profits than if they were just serving the urban market. 4 More generally, differential serving areas and COLR obligations create strategic incentives which will influence the level of competition between carriers. Theoretical work has thus far been
constrained to the case of an incumbent competing with a new entrantthe case
of competing existing COLRs has not been modeled. Strategic considerations and information asymmetries have yet to be analyzed in this environment.
The next question is whether these areas should be large or small. When there
are potentially significant cost complementarities (costs depend on the specific combination of areas that a service provider will serve), then there are two options: (i)
auction a large enough areas to include most of the significant complementarities;
or (ii) auction many smaller areas, but permit for combinatorial bidding so that significant complementarities can be realized. There appears to be some dispute about
the feasibility of (ii) [Kelly and Steinberg (1998) claim that complex combinatorial
auctions are feasible, but Hultkrantz (2004) cites Kelly and Steinbergs work, but
concludes that the consensus in the economic literature seems to be that
combinatorial auctions have several desirable properties but are too difficult to be
used; Sorana (1998) claims it must be ultimately recognized, however, that the
theoretical and experimental properties of multi-unit auctions, combinatorial or otherwise, are not well understood, and Luander and Nilsson (2004) provide experimental evidence that combinatorial auctions may be more efficient and make collusion more difficult than one shot sealed auctions].
Large area auctions would appear to favor larger carriers, or would require smaller carriers to bid jointly in order to compete.5 Larger areas that make sense from
a network perspective may also require a mixture of areas currently served by rural
and nonrural carriers. This would exacerbate the complexity of designing joint bids
to serve large areas. It may also increase the size of the Fund by including highcost areas (currently served by nonrural carriers) that do not presently receive support.
In general, smaller areas should involve more precise and larger Universal Service funds, ceteris paribus. Larger areas involve more averaging of relatively high
and relatively low cost customers, tending to decrease the overall fund size, but failing to provide full support for high-cost areas [Lehman (2000)]. Smaller areas necessarily involve the complexities of combinatorial bidding.
The averaging effect can be substantial. At the extreme, imagine a single national
service area being auctioned offa subsidy would probably not be required to serve
the high-cost areas along with the low cost areas. This result, however, is a move
away from decades of efforts aimed at increasing competition in the industry. If auctions are designed to accommodate large areas and competition within these areas,
then the overall Fund cost will be driven upwards, as discussed below under the
number of COLRs.
Determination of geographical areas to be auctioned is complicated by the presence of multiple existing network infrastructures. For example, suppose that the
COLR includes service to 100 percent of the customers within a current ILEC serving area and that a wireless carrier wishes to bid, but their network only covers
80 percent of the population in that area. The wireless carrier would be required
to arrange to resell the incumbents service or provide an alternative infrastructure
for the 20 percent of customers that it does not currently reach.
Conversely, suppose the service area is defined as the wireless carriers service
area, and that this extends beyond any single ILECs service area. This would require several ILECs to combine their bids to match the service area of the wireless
carrier. In either case, transactions costs and uncertainty will increase when existing infrastructures do not match.6
4 Hoernig

and Valletti (2003) at page 91.


5 Current spectrum auctions highlight this issue. Joint bidding is permitted, but the bidders
cannot subsequently use the spectrum rights individually, under their separate business identities. Auction design should avoid dictating market structureit should reveal when joint bidding is most efficient, but it should not force carriers to consolidate operations. Forced consolidation presupposes that regulators know the most efficient market structure to begin with, undermining the potential of auctions to substitute market processes for regulatory processes.
6 The 1999 NPRM cited the use of competitive bids for COLR in Hawaii. The first such award
went to TelHawaii. In order to transfer the assets from the previous COLR, GTE Hawaiian Tel,
the Public Utilities Commission of Hawaii condemned some of the assets of GTE Hawaiian Tel.
Several court battles later, a state court overturned the condemnation as unconstitutional. Rather than continue the legal battles, TelHawaii pulled out of the market after spending millions
of dollars attempting to enter [Honolulu Star-Bulletin, July 20, 1999]. Regardless of the ultimate

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00066

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

63
It is difficult to design an auction that will be technologically neutral under these
circumstances. To avoid bias, areas would need to be smaller than anybodys current
service area, thereby placing a similar burden on all potential bidders. However,
such small areas would greatly increase the complexity of the combinatorial auctions that would be required.
Number of COLRs
Closely related to defining the geographic COLR area is the issue of whether
there will be one winning bid or more than one within each area. At a fundamental
level, there is a tradeoff between competition for the market (favored by a single
winning bidder) and competition within the market (promoted by multiple winning
bidders). A priori, it is not clear which type of competition would lead to greater
economic efficiency.
It is clear that total subsidies will be larger with multiple winning bids than single winners. This is evident from the GTE reverse auction proposal submitted to the
FCC [Weller (1998)]. Weller proposed that bidders submit two bidsone for sole
provision of COLR within an area and the other assuming shared provision of
COLR responsibilities. Preliminary evidence was that reducing a carriers market
share by 50 percent would increase unit costs by 52 percent. This is due to the fact
that network investment is not proportional to the number of customers, particularly in sparsely populated areas. Serving half of the customers may entail nearly
the same infrastructure as serving all of the customers.
It should be noted that some technologies may be more tolerant than others of
multiple winning bidders. Wireless technology does not have the same sunk cost
characteristics as wireline technology, so per unit subsidies may not increase as dramatically for wireless carriers. This need not cause a problem as long as the
wireline bidder can receive a subsidy adequate to serve a partial market share. If
high-cost support is capped at current per-subscriber levels, adequate support would
be impossible, however. So, it is important that there be no caps on bids if multiple
COLRs are to be awarded.
Single COLRs does lead to reduced USF costs in one wayit eliminates the problem of multiple supported services (wireline and wireless) without the administrative problems that accompany proposals to limit individual support to a single service (to households, or locations, etc.).
Sorana (1998b) examines an auction mechanism (based on the 3rd lowest bid)
that permits multiple COLRs. He points out that there could be much higher cost
involved if the auction rules are not carefully crafted. This results from the vulnerability to collusion. While careful auction rules can avoid this (by making the number of COLRs dependent on the bid amount) it may still be unable to generate
enough incentives for high-quality service.
Laffont and Tirole (2000) provide an extended theoretical analysis of reverse auctions, focused principally on the issue of multiple COLRs. They conclude:
We are unaware of formal analyses of Universal Service auctions with endogenous market structure. We have tried to provide a framework within which
analysis of such auctions can begin. The first insights thus gleaned do not build
as strong a case for the introduction of competition as we had expected. 7
One salient point is that endogenous market structure increases uncertainty for
bidders, thereby requiring an extra risk premium in their bids. Laffont and Tiorle
also echo the complexities raised by existing infrastructure in high-cost areas,
Much of the discussion on Universal Service auctions proceeds as if all competitors were building their network from scratch. This may be a fine assumption for newly settled areas or when substantial network upgradings are contemplated. In practice, however, many high-cost areas are already partly covered by a wire-based incumbent operator able to provide the supported services
with its existing technology. While the incumbent operators network may have
been very costly to build, once in place it has a low (short-term) marginal cost.
And so facilities-based entrants (e.g., offering wireless services) may find it hard
to compete with the incumbent. In our view, more attention should be devoted
to this aspect of Universal Service provision. 8
In the U.S. environment, the issue is doubly complex since there is existing wireless infrastructure in many high-cost areas. The theoretical performance of auctions
merits of the legal dispute, problems like this are likely to accompany bids that require use of
other carriers facilities in order to satisfy the COLR obligations.
7 Laffont and Tirole (2000) at pages 254, 260.
8 Laffont and Tirole (2000) at page 260. This point was also made by Milgrom (1996).

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00067

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

64
has not yet been studied under these circumstances. Nor is there much empirical
evidence to provide guidance.
Duration
There is a tradeoff between long and short duration of COLR franchises. Short
time periods enhance the ability of Universal Service costs to adjust to changes in
technology or changes in service definition. However, this comes at the cost of inhibiting investments that have longer time horizons.
It is notable that cable franchise awards (where competitive bidding is used) are
quite longtypically 815 years. It is difficult to reject a renewal application upon
expiration. Federal law places the burden of proof for failing to renew a cable franchise on the communitythey must show that the carrier is either unable to continue providing the service or will be unable to provide the service that the community requires in the future [Kramer (2003)]. In fact, in the 1980s, only 7 out of 3,516
cable refranchising decisions resulted in replacement of the existing franchise owner
[Zupan (1989)].
There is a relationship between contract duration and the number of winners.
Even with single auction winners, issues arise concerning whether the incumbent
winners should have any special treatment in subsequent auctions, or whether there
are benefits to opening future auctions to carriers other than the prior winners.
Laffont and Tirole (2000, page 261) reach the conclusion that,
the incumbent may be shut out of the market. The transfer of the incumbents
capital to winning entrants (either through rentals or through an acquisition)
may give rise to the usual concerns about the impact of second sourcing on
the incumbents incentives to invest in the quality of its network.
Previous work by Laffont and Tirole (1988) explored the case where incumbents
investments are observable (i.e., where they can be acquired by othersan example
of unobservable investment is the buildup of knowledge within the human capital
of the firms managers: it seems that most rural incumbent investment is observable, such as the physical capital of the infrastructure). They reach a relatively pessimistic assessment of the virtues of second-sourcing (or takeover) when substantial
investments are at stake. (page 532) This is due to the potential that some of the
value of the incumbents investment may flow to future auction winners. This externality causes the incumbent to under-invest, and calls for future auctions to be
stacked in the incumbents favor. Indeed, this is a rationale behind the burden of
proof in cable refranchising that falls on those that do not want a franchise renewed.
Universal Service minimum subsidy auctions in South America have typically
used lump-sum payments with 5 year exclusive franchises [ITU (2002)]. The subsidy
is paid in stages, according to established milestones (e.g., upon installation of half
of the required payphones), but it is not a recurring payment. That is, the subsidy
is geared to recover the full cost of the investment (unless the bidder is willing to
bid for only partial recovery during the 5 year period). Carriers can decide how
much risk they wish to bear by bidding for less than full recovery during the 5 year
period. Given that these South American auctions (and new ones proposed in Africa)
take place in green-field environments, there is often a business case for ultimate
expansion into these unserved areas, so bidders may be willing to accept less than
full cost recovery from the subsidy mechanism. It is unclear how relevant these circumstances are to the U.S. rural environment (where many rural areas are not
growing).
Sorana (1998) points out that sufficiency of USF is not assured by good auction
design, and neither is voluntary provision of universal service. He constructs a
model to compare reverse auctions with cost-proxy models, finding that auctions
may involve lower subsidies than accurate cost proxy models, but his model assumes
that the funds from the auction are sufficient for the intended purposes. He notes
that this is not assured.
Competitive bidding is used in the Essential Air Service program, but with only
a 2 year horizon. Airplanes, however, are quite mobile, unlike telecommunications
infrastructure. These examples suggest that the time periods would have to be relatively long, if there is to be sufficient incentive to invest in telecommunications infrastructure.
Transition
Existing infrastructure complicates the picture. Suppose the incumbent loses the
auction but has investment that was prudently incurred, but has not yet been fully
recovered. It is possible that the winning bidder may want to purchase this infra-

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00068

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

65
structure.9 This creates legal and policy issues, but it also impacts economic efficiency. If regulators establish a precedent for truncating recovery of prudent past
investments, then future investment will be affected. It is unlikely many investments will take place with payoff periods longer than the duration of the franchise.
The World Bank (2000, pages 626) cites competitive bidding as a feature of a
good universality fund, but As previously discussed, the process is more difficult
where an incumbent is already providing the designated universal services. The
embedded network may provide the incumbent with an advantage bidding against
new entrants (as was the case in India and Australia, discussed below), or may force
the incumbent to fail to recover its past investments, despite regulatory oversight
deeming those investments to be prudent.
Despite these complications, the World Bank does claim that auctions are still
possiblethey cite transfer of assets to the lowest bidder, subcontracting, joint ventures, etc. as mechanisms that can deal with embedded infrastructure. While such
developments can enhance efficiency, there are costs associated with each of these
avenues (as demonstrated in the Hawaii case in footnote 6).
The only way to avoid bias either for or against incumbent networks is to fully
recover the incumbents investment prior to enacting the reverse auction. It is not
surprising that the most successful reverse auctions (Chile, Peru, Guatemala, Columbia, and the Dominican Republic) involved previously unserved areas or significant upgrades to the existing infrastructure within these areas [ITU (2004)].
The need to address stranded investment is well-recognized in the area of electricity deregulation. The Congressional Budget Office (1998) reviewed the stranded
cost issue, concluding,
For reasons of fairness and political reality, utilities are likely to be compensated for some or all of their losses. Determining the correct figure for
stranded costs, deciding how much of them to compensate, and figuring out how
that compensation should be paid are difficult issues, which are slowing
progress toward restructuring in many states.
Volumes have been written and disputes continue over measurement and recovery
of stranded electric generating costs, but it is an issue faced by all attempts at deregulation.
For example, in Texas, there is a provision for true-up charges:
These true-up proceedings are designed to provide commission authorization
for an electric utility to begin recovery of its costs for power plants built to meet
customer demand for electricity prior to the start of retail competition, which
cannot be recovered in the competitive marketplace. These costs are said to be
stranded. 10
Reverse auctions potentially render the incumbents network less valuable (if they
lose the bid or forego full cost recovery in order to win the bid). Given that these
were prudent investments undertaken precisely to fulfill the COLR, there is a
strong case for recovery of these stranded costs. To the extent that new technologies
(e.g., wireless) cause this decrease in value, the case for recovery is strengthened
(since the investments were prudent at the time they were made, and were often
recovered through overly long depreciation schedules). Resolution of this issue is of
political, legal, and economic importance (the latter through its affect on future investment incentives).
Eligibility
Bidders must be financially and operationally capable of fulfilling their COLR responsibilities. The FCC has considerable experience with ensuring bidder eligibility,
although there have been problems, particularly with small bidders. The goal should
be to have enough bidders to ensure a competitive bidding process, while limiting
future problems with failure to deliver the required services.
The 1999 Peru auctions illustrate this problem [ITU (2004)]. The winning bid was
20 percent of the available subsidy, but the winning company then could not meet
its targets. The ITU presents this an example of excessively low bidding and points
out that most Latin American auctions have attracted bidders without much operational experience, and have failed to attract large international operators or incumbents.
9 Although this may entail problems such as those encountered in the Hawaii case discussed
above.
10 Described at http://www.aep. com/newsroom/resources/docs/TrueUp.pdf#search=%22stran
ded%20investment%20auctions%22.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00069

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

66
Summary on Geography, Size, Numbers, and Eligibility
The discussion thus far can be summarized as a spectrum of choices that would
govern the intensity of competition for the COLR subsidy. International experience
can be placed on a continuum from lack of competition to healthy competition. The
Latin American examples [World Bank (2000), ITU (2002), ITU (2004), Intelecon
(2005), Scherf (2006)] appear to have had truly competitive bidding in their reverse
auctions. Savings of 50 percent (compared with the maximum potential subsidy
level) are commonly cited, but these savings are based on comparison with a cost
proxy model of unknown accuracy. There is no evidence concerning the relative costs
of reverse auctions and other Universal Service mechanisms in any of these countries. Still, the auctions were administratively feasible and resulted in multiple bidders for the COLR.
The extreme example of a lack of competition for the market is India [Malik and
Silva (2005), Noll and Wallsten (2005)]. Reverse auctions were held for infrastructure upgrades to a number of rural areas. The incumbent, BSNL, won almost all
of the bids and bid the maximum subsidy available in each case. Critics of the Indian auction point out that the eligibility rules essentially predetermined this outcome. Only providers with current infrastructure in these regions could bid; technologies were limited to wireline and fixed wireless, and bidders were required to
install infrastructure to reach everyone within these regions but without any wholesale regulation of the incumbent to provide for interconnection, unbundling, or resale. As a result, in 19 of the 20 areas, there was only a single bidder (BSNL) and
they bid the maximum subsidy available. The rules were designed to promote neither entry nor efficiency.
The other end of the spectrum can be envisioned as the U.S. While competitive
bidding has not been utilized, support on a predetermined per line basis (i.e., without uniform coverage requirements) has been offered to multiple ETCs. The fact
that many rural areas have witnessed multiple carriers willing to accept the offered
support level, suggests that there would be multiple bidders if the auction were conducted on a per-line subsidy level, and without requirements to serve everybody
within the same service areas with the same quality characteristics. In this sense,
the current rules for the high-cost fund are designed to promote entry, but not efficiency.11
Australia provides an interesting data point [Department of Communications, Information Technology and the Arts, Australia (2004), ITU (2006)]. Two pilot regions
were selected for reverse auctions. These included the most remote 80 percent of
Australia, and $150 million was available for introducing unlimited local calling
with these areas. The goal was to find a simpler way of determining a reasonable
level of subsidy de-linked from a calculation of costs. 12 The auction was designed
for a single winner. No competitive tenders were received. In fact, since 1991, carriers other than the incumbent (Telstra) have been free to apply to be COLR, but
none have applied. The ITU report concluded However, while the experiences with
designating Universal Service providers on the basis of competitive tendering in
some countries has been encouraging (e.g., Chile and Peru), there has been some
less positive experience in Australia.
Australian regulators did follow-up analysis to determine the causes for lack of
competitive interest. Major factors cited were: difficulty competing with Telstra,
meeting the obligation to serve all customers, and difficulty identifying other revenue opportunities to help support COLR responsibilities. It is also possible that the
investment climate at the time of the pilots was unfavorable. The regulator concluded that higher subsidies might induce entry, but they were not worth the significant increase in costs. They recommended preserving the reverse auction option,
but not continuing it at this time. One benefit they cite from the pilots is the determination that Telstra was not being overcompensated for COLR at current subsidy
levels.
Another example is provided by electricity deregulation in Maine [Maine Public
Utilities Commission (2002)]. Maine claims to have the most robust retail competition for electricity customers in the Nation. Significant competition (more than half
of the market) has developed for large customers. Virtually no retail competition has
developed for small residential and business customers (with the single of exception
11 Parties differ in the source of inefficiency that they see, but virtually all agree it is inefficient. Some parties point to the support of multiple carriers based on incumbent costs as leading
to unnecessary duplication of infrastructure and unnecessary support for CETCs. Others believe
the waste is caused by the cost plus nature of determining support levels. In any case, nobody
claims the current environment is particularly efficient.
12 ITU (2006) at page 14.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00070

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

67
of a small area in northern Maine, which the Commission discounts for a number
of region-specific reasons).
State legislation eliminated the obligation to serve, with standard offer service
available for those who could not find a suitable competitive supplier. The Commission was instructed to strive for at least 3 suppliers of standard offer service in
every areas, but only if multiple suppliers would not cause rates to be significantly
higher.
Early attempts to solicit competitive bids for standard offer service did not result
in retail suppliers for all customer classes. Later attempts were somewhat more successful. Still, the Commission notes that there is virtually no retail competition for
residential and small commercial customers, either in Maine or elsewhere. Their
research concludes that prices should not be increased in the hope of attracting suppliers (consumer input was strongly against paying higher prices in exchange for increased competition). Standard offer service does extend some of the benefits of competition to individual small customers through the aggregation inherent in a standard offer available throughout the state. In the telecommunications context, this is
akin to requiring geographical averaging of retail prices across broad geographic regions. This is closer to the old system of implicit support in which lower cost customers pay higher prices in order to support lower prices for the high-cost customers. Such a system is not feasible in a truly competitive environment.
What these examples reveal is that regulators have wide discretion in determining the extent of competition for the market that results from a reverse auction.
They can design auctions that preclude entry (such as in India) or they can promote
entry, regardless of attendant inefficiencies (the U.S.). It appears to be feasible to
get reasonable entry and efficiency in a green-field environment. This is what the
Latin American examples show. It is more elusive in environments with existing
providers.
The political economy of regulatory policy must be considered when evaluating reverse auctions. In the absence of strong policy direction, it will be difficult to design
a reverse auction that does not either deny CETCs their current support or deny
rural ILECs recovery of their existing investments. The result could well be a managed competitive reverse auction, with few of the benefits that reverse auctions potentially offer.
To avoid a managed outcome, regulators must set a clear goal in terms of how
much entry they want, and what efficiency cost they are willing to bear. A concrete
example is the choice of serving area. Very small geographical areas can promote
entry (per-subscriber subsidy bids is the extreme example), but jeopardize the ability to realize cost complementarities and at the risk of unnecessary duplication of
support. The trouble is that regulators must know a great deal about what is most
efficient before they can design the reverse auction (for example, they must know
how many COLRs are efficient, and which technologies are most efficient, and how
to define Universal Service over the length of the franchise contract). It is the absence of such knowledge that is one of the major benefits of using reverse auctions
to begin withthe market is supposed to provide these answers.
It is the existence of current infrastructures that complicates this design. Rules
cannot be chosen that will satisfy all interests, so the regulator is required to know
what the efficient outcome looks like before the auction can be designed. In a greenfield environment, by definition the COLR that is being auctioned is one that the
market has not found profitablehence, there are fewer interests at stake in the
creation of the reverse auction mechanism. The evidence supports this conclusion:
green-field reverse auctions have been fairly successful, while there are no clear examples of competitive bidding in more developed settings.
Auction Mechanics
There are a number of subsidiary design questions that deal with the mechanics
of how a reverse auction would actually operate.
Type of Bidding
Most reverse auctions have utilized simple one-shot sealed auctions. Most spectrum auction design has been multiple-round, open, combinatorial auctions. The underlying issues concern the importance of cost/value complementarities, bidder risks,
and opportunities for collusion. These have been extensively studied in the general
auction literature. A few particular considerations apply in a Universal Service setting. Cost complementarities are potentially important, so the auction must either
be combinatorial or involve fairly large geographical areas. Both pose problems. In
addition, in an environment in which there are existing infrastructure providers,
sealed bidding would appear to impede much necessary negotiation about joint bids,
outsourcing arrangements, etc. Some research suggests that sealed bidding may ac-

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00071

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

68
tually facilitate collusion [Luander and Nilsson (2004)]. On balance, it would appear
that combinatorial bidding is more appropriate in the U.S. environment, but the feasibility and complexity of the required auction is in some dispute.
Determination of Winner(s)
It is clear that more than price must be considered in determining winning bids.
None of the international examples (or domestic examples from other industries) entail a price-only selection. What the literature does say, however, is that the rules
for determining winners must be specified precisely and unambiguously in advance
[ITU (2002), World Bank (2000)]. That is, the process must avoid subjectivity. This
is the same problem encountered in many procurement contractsthe rules must
be clear and objective.
Current costs, under the U.S. high-cost fund, are controlled via a number of oversight mechanisms, the lack of full cost recovery (high-cost funding only supports a
percentage of the costs above the national average), and competitive pressure from
other services (e.g., VoIP, wireless usage substituting for wireline usage, etc.). The
high-cost fund, itself, is not designed to necessarily minimize costs. It does not contain cost-reducing incentives as strong as would an auction mechanism. While this
can lead to inefficiency in terms of costs, it also permits more flexibility in terms
of services offered (e.g., broadband speeds). This flexibility has valueparticularly,
if regulators do not have sufficient information to project Universal Service definitions into the future.
Post-Award Flexibility
Reverse auctions in developing countries have relied on additional service revenues to reduce the cost of public subsidies. Permitting COLRs to market valueadded services, in addition to the contracted COLR, can result in their bids being
less than the cost of providing solely the COLR. Some countries have specifically
permitted retail prices in rural areas to exceed those in urban areas by predetermined amounts. In some auctions (e.g., the Essential Air Service Program) there are
no restrictions on post-award pricing at all.
It is clear that bidders will bid lower in a reverse auction to the degree that they
have post-award flexibility. However, flexibility endangers the concept of universal
service. Once again, there is a tradeoff. The more flexibility that is provided, the
lower the expected subsidy required, but the less assurance there is that Universal
Service objectives will be met.
It is also worth noting that the successful Latin American reverse auctions rely,
in part, on asymmetric interconnection fees to support rural providers. For example,
the largest Chilean rural operator gets 60 percent of its total revenues from such
charges; Columbia has recently introduced asymmetric fees, and Peru plans to [ITU
(2004)]. They also permit higher rural prices and lower license fees in rural areas.
Uganda has recently introduced a reverse auction for service to 154 communities
that no operators were willing to serve, and part of the mechanism was permitting
voice service rates in rural areas to be up to 50 percent above rates in Kampala
(as well as higher termination fees in rural areas) [Intelecon (2005)].
Monitoring and Enforcement
Performance under the franchise award must be monitored. Most countries have
specific penalties for failure of winning bidders to meet their performance targets.
Removal of a COLR, either through failure to perform adequately or through carrier
bankruptcy, poses particular problems for reverse auctions for universal service.
How is service to be guaranteed for rural customers in the event that their winning
bidder does not (or is unable to) meet its obligations? Scherf (2006) cites this as a
weakness in the build-out requirements that accompany licenses in many developing countries: it is cheaper to pay the penalties than fulfill the requirements.
Bankruptcy risks are somewhat mitigated under the current USF by the historic
regulatory compact in which rural ILECs have been able to recover their past investments. When cost recovery becomes more uncertain, and when awards are based
on low subsidy bids, these risks increase.
Scherf (2006) says that the regulatory environment has to be credible and sustainable to the eyes of investors, (page 12) and discusses issues associated with enforcement mechanisms, particularly in developing countries. He cites problems in
Peru, where some very low bids had been submitted, with subsequent renegotiation
under the threat of carrier bankruptcy. He also mentions Uganda, where the regulator has not even asked for the performance data it would need to monitor performance. These concerns are more pronounced in countries with less developed political
institutions, but they also arise in the U.S. In addition, we have the issue of the
appropriate jurisdictional responsibility for monitoring and enforcement.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00072

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

69
Conclusions
In a definitive work on the theory and practice of auctions, Klemperer (2004) concludes,
In conclusion, the most important features of an auction are its robustness
against collusion and its attractiveness to potential bidders. Failure to attend
to these issues can lead to disaster. And anyone setting up an auction would
be foolish to blindly follow past successful designs: auction design is not one
size fits all. . . . In the practical design of auctions, local circumstances matter
and the devil is in the details. 13
Auctions have a number of desirable properties. The ITU states that,
The use of well-designed competitive tenders can (in certain circumstances)
help to generate incentives to contain costs, innovate, and reveal the true cost
of delivering Universal Service (thus helping to minimize the subsidy required. 14
We have seen that auctions can be feasible and effective for provision of Universal
Service in unserved areas, if they are properly designed. Their success depends on
an appropriate definition of the objective for universal service. Reverse auctions
have been most successful where the objective can be clearly defined and does not
require long-range forecasting: e.g., provide payphone service in specified rural villages (Chile, Peru, Columbia, Guatemala).
Reverse auctions in the U.S. are a different matter. There are multiple existing
infrastructures, utilizing different technologies, providing different services, and
with different serving areas. Universal Service is an evolving set of service requirements that is difficult to forecast. The performance of auctions in this setting is
theoretically and empirically untested. The limited evidence suggests that these are
difficult problems.
Auction design will need to address competition within the market as well as for
the market, potentially large cost complementarities between high-cost areas as well
as between high-cost and low-cost areas, and provide for investment incentives with
significant sunk costs and technological uncertainty.
Much of the theoretical appeal of reverse auctions is dissipated under the actual
conditions under which Universal Service will be provided. Regulators will need
more foresight than they would like. They will need to specify Universal Service requirements far enough into the future to allow for the required investment incentives. They will need to know more about the most efficient market structure (single
COLR, multiple, which technology, etc.) than they would like. Auctions are supposed
to permit the market to make these determinations, not regulators. But, this benefit
can be illusive. Can the market pick the technology if the auction design cannot put
different technological platforms on an equal footing?
One clear beneficiary of a reverse auction system is the economics profession.
Their expertise lies in auction design and the devilish details contain plenty of interesting work. How consumers of Universal Service and providers will fare, is less
clear.
The Joint Board Discussion Proposal
The Discussion Proposal (The Proposal) provided with the Joint Board Public Notice provides a good illustration of the difficulties of applying reverse auctions in a
nongreenfield environment. The Proposal does not appear to be derived from any
theoretical efficiency properties, nor does it follow the reverse auctions that have
been implemented elsewhere. Instead, it seems to be driven by the need to accommodate the fact that we are currently supporting multiple networks using multiple
technologies in rural areas.
Separate support for broadband and mobility services in rural areas for 10 year
periods, takes a particularly static view of technology. It provides support to two
sets of services, neither of which are included in the current definition of universal
servicemobility and broadband. The Proposal does attempt to address the transition issue by offering an initial phase-in whereby rural ILECs can elect to receive
support (at current levels plus inflation) for the first 10 year period for broadband
service. This is recognition that past prudently incurred investments need to be recovered.
But, what happens after 10 years? What will govern future network investment?
Here, the Proposal is silent on the details that will ultimately determine future Universal Service in rural America. The Proposal says that ETCs would be required to
13 Klemperer
14 ITU

VerDate Nov 24 2008

14:09 Mar 21, 2012

(2004) at page 122.


(2006) at page 25.

Jkt 073387

PO 00000

Frm 00073

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

70
relinquish essential facilities at fair market value at the end of the contract term.
After 10 years of trying to determine fair market value for unbundled network elements under the Telecommunications Act of 1996, the task of determining fair market value for essential rural network facilities will be daunting.
The Proposal defines geographical coverage as 90 percent or more of the households, without specifying how ETCs would acquire the services needed to reach the
remainder of the households (echoing some of the problems in the Australian and
Indian reverse auctions discussed above). Basic geographical units would be counties, with the exception of rural ILECs, and counties could be bid on in bundles or
separately. This does not address the complexity of the combinatorial auction that
would be required (the U.S. Census Bureau lists 3,141 counties or county-equivalent
administrative units), nor does it address the issue of whether the mobility support
would extend to all counties, including those served by nonrural ILECs. There is
the potential for a significant growth in the fund, if it includes currently unsupported areas.
Upon review of the past successes with reverse auctions, they appear to deliver
tangible benefits when used to support delivery of services where current infrastructure is not in place. While many rural areas see significant competition among wireless carriers, there is still a need for more extensive build-out of rural networks. The
mobility USF could be aimed at this goal, by tying support to specific infrastructure
targets.
The Proposal illustrates the complexity of applying reverse auctions in the existing mixed technology infrastructure of the United States. The devil is in the details,
but the details are not in the Proposal.
References
Australia, Department of Communications, Information Technology and the Arts.
2004. Review of the Operation of the Universal Service Obligation and Customer
Service Guarantee.
Congressional Budget Office. 1998. Electric Utilities: Deregulation and Stranded
Cost.
Demsetz, H. 1968. Why Regulate Utilities? Journal of Law and Economics, Vol.
11, 5566.
Goldberg, Victor. 1976. Regulation and Administered Contracts. Bell Journal of
Economics, Vol. 7, 426452.
Hoernig, Steffen H. and Tommaso M. Valetti. 2003. The Interplay between Regulation and Competition: The Case of Universal Service Obligations. Chapter 6 in
Spectrum Auctions and Competition in Telecommunications, edited by Gerhard Iling
and Ulrich Kluh, The MIT Press.
Hultkrantz, Lars. 2004. A fresh start on universal-service policy. Paper presented
at the Swedish National Post and Telecom Agency Conference.
Intelecon. 2005. Universal Access and Universal Service Funds: insights and experience of international best practice.
International Telecommunications Union (ITU). 2002. Model Universal Service/Access Policies, Regulations and Procedures. Part II: Minimum Subsidy Competitive
Auction Mechanisms for Funding Public Telecommunications Access in Rural Areas
and Tariff/Interconnection Regulation for the Promotion of Universal Service/Access.
Paper prepared by Edgardo Sepulveda for the Global Symposium for Regulators,
Hong Kong, December 78, 2002.
International Telecommunications Union (ITU). 2004. Leveraging Telecommunications Policies for Pro-Poor Growth Universal Access Funds With Minimum-Subsidy Auctions. DAC Network on Poverty Reduction, Development Co-operation Directorate, Development Assistance Committee.
International Telecommunications Union (ITU). 2006. What Rules for Universal
Service in an IP-Enabled NGN Environment? Background Paper for ITU Workshop, Geneva, March 2324, 2006.
Kelly, Frank and Richard Steinberg. 1998. A Combinatorial Auction with Multiple Winners for Universal Service. Subsequently published in Management
Science: 46(4), 586596, 2000.
Klemperer, Paul. 2004. Auctions: Theory and Practice. Princeton University Press.
Kramer, Jonathan. 2003. Leveling the Playing Field for Cable-TV Franchise Renewals. Public Management; 85, 11.
Laffont, J.J. and Jean Tirole. 1988. Repeated auctions of incentive contracts, investment, and bidding parity with an application to takeovers. RAND Journal of
Economics, Vol. 19, No. 4.
Laffont, J.J. and Jean Tirole. 2000. Competition in Telecommunications. The MIT
Press.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00074

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

71
Lehman, Dale E. 2000. Who Will Serve Rural America? white paper for the National Telephone Cooperative Association.
Luander, Anders and Jan-Eric Nilsson. 2004. Combinatorial Procurement AuctionsA Collusion Remedy? Unpublished Manuscript.
Maine Public Utilities Commission. 2002. Standard Offer Study and Recommendations Regarding Service after March 1, 2005.
Malik, Payal and Harsha de Silva. 2005. Diversifying Network Participation:
Study of Indias Universal Service Instruments. WDR Dialogue Theme 3rd cycle,
Discussion paper WDR0504, regulateonline.org and LIRNEasia.
Milgrom, Paul. 1996. Procuring Universal Service: Putting Auction Theory to
Work. Lecture at the Royal Swedish Academy of Sciences.
Noll, Roger G. and Scott Wallsten. 2005. Universal Telecommunications Service
in India. AEI-Brookings Joint Center, Related Publication 0525.
Scherf, Thorsten. 2006. Policies for universal Access to Telecommunications in
Rural Areas of Developing CountriesA comparative Analysis. Unpublished paper,
available at http://zeus.econ.umd.edu/cgi-bin/conference/download.cgi?dblname=I
IOC2006&paperlid=544#search=%22scherf%20universal%20service%22
Sorana, Valter. 1998a. Some Economics of Carrier of Last Resort Auctions.
Paper presented at the Telecommunications Policy Research Conference.
Sorana, Valter. 1998b. Auctions for Universal Service Subsidies. Unpublished
manuscript, Stanford University.
Weller. 1998. Auctions for Universal Service Obligations. Paper presented at
12th biennial conference of the International Telecommunications Society.
Westerveld, R. and C.F. Maitland. Technical and policy advances in rural telecommunications. Unpublished manuscript.
World Bank. 2000. Telecommunications Regulation Handbook: Module 6, Universal Service. Edited by Hank Intven, McCarthy Tetrault.
Zupan, M.A. 1989. Cable franchise renewals: do incumbent firms behave
opportunistically? RAND Journal of Economics 20, 473482.

The CHAIRMAN. Thank you very much, Mr. Crothers.


May I now recognize Dr. Staihr?
STATEMENT OF BRIAN K. STAIHR, PH.D., REGULATORY
ECONOMIST, EMBARQTM CORPORATION

Dr. STAIHR. Thank you very much.


Im Brian Staihr. Im an Economist for Embarq. Very happy to
be here today.
Now, Embarq is the countrys largest independent wireline telephone company. We have about 7 million customers across 18
states. And if you look at the picture that Brians showing you over
here we serve some very wonderful rural areas, like Possum Kingdom, Texas, and Pretty Prairie, Kansas.
[Laughter.]
Dr. STAIHR. And because were very rural, we appreciate the time
and the effort that this committee has put into the subject of Universal Service. Already this year, Senator Stevens has introduced
the USA Act, which addresses many important issues. And we look
forward to working with all of you in the future.
Now, quickly to reiterate two facts, we all know that the Federal
Fund has grown significantly in the past few years and the FCC
is looking at ways to control this growth, including auctions. We
also know, if you look at this graph, that the source of the growth
is receipts that have gone to competitive carriers rather than to incumbents that serve as carriers of last resort. Now, this difference
is significant, because when a company is a carrier of last resort,
it has an obligation to serve all the customers in an area, including
the high-cost customers that nobody else wants to serve. Competitive carriers and wireless carriers dont have this obligation even
when they get USF.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00075

Fmt 6633

Sfmt 6601

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

72
Now, to see why this difference is significant, I want to show you
a picture of Meadowview, Virginia. The different colors in the picture on the left represent different population densities. Yellow and
red show high density, which is low cost. The green shows low density, which is high cost. Embarq, as the carrier of last resort, has
to serve the whole thing, the yellow and the green parts.
The second picture shows you the wireless coverage in
Meadowview. As you can see, the wireless coverage pretty much
stops where the high-cost parts start. The wireless carrier doesnt
have an obligation to serve the high-cost area; and, even if it receives USF, it doesnt have this obligation.
Now, this picture illustrates a key problem with the Fund as it
exists today. Before we had competition, a company like Embarq
would serve Meadowview, and we could count on the low-cost areas
offsetting the high-cost areas. If we lost money serving the green
part, that was OK, because we served the yellow part, too, and, on
average, we were all right. But, after 1996, competition developed,
and, in a place like Meadowview, it developed just in the yellow
areas. As a result, we could no longer count on that low-cost offsetting the high cost, because wed lost half the customers in the lowcost area.
The point here is, the Federal Universal Service Fund has not
kept pace with this competitive reality, because, when the current
system looks at a place like Meadowview, it assumes that Embarq
can continue to use the low-cost areas to offset the high cost, and
we cant do that anymore.
In addition, under the current system, this competition creates a
very strange kind of chain reaction. Competitors come into the lowcost areas, they get the same support per line as the incumbent.
This support draws more competitors into those same areas. That
means the competitors serve low-cost customers, the incumbent
serves high-cost customers, the incumbents costs go up, the support goes up, we end up oversupporting the town center and basically shortchanging the outlying areas.
The way to fix this is to target support more granularly, to reexamine the area that we look at when we determine the need for
Universal Service support, particularly to consider the town center
and the outlying areas differently.
What will this do? Three things. First, itll stop that chain reaction. Second, itll target the support to where its really needed.
And, third, it will eliminate this reliance on these unsustainable
cross-subsidies, while not necessarily increasing the size of the
Fund.
Now, Ive got one more picture to show you. This is Fort Meade,
Florida. Every green dot on that picture is a customer location. You
can see theres a very clear downtown area. That area is pretty
low-cost. The outlying areas are much higher-cost. All right? The
outlying areas dont see competition, in general. When we see competition in Fort Meade, its, just like Meadowview, in that downtown area. As a result, the outlying areas cant be subsidized by
the downtown. The outlying areas need support. And, under the
current system, they dont get any.
Targeting USF would bring rationality to the USF distribution
system. And its not mutually exclusive with other policy consider-

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00076

Fmt 6633

Sfmt 6601

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

73
ations that were looking at today. We can talk about reverse auctions. We can talk about support for broadband. We can talk about
eliminating identical support. We can talk about more granular
support, in conjunction with any of those. Or we can talk about
more targeted support, apart from any of those. It works both
ways.
Now, to wrap things up, 11 years ago when the Act was passed,
we didnt have much competition in rural America, we didnt have
the capabilities or the tools to calculate support specifically for
these outlying areas. Today, we have the capability, we have the
tools, and we have one more thingwe have the incentive, going
forward, to do it right.
So, with that, Ill stop. I appreciate the time today and look forward to any questions you may have.
[The prepared statement of Mr. Staihr follows:]
PREPARED STATEMENT

OF

BRIAN K. STAIHR, PH.D., REGULATORY ECONOMIST,


EMBARQTM CORPORATION

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00077

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

301STAI1.eps

Good morning, Mr. Chairman, Vice-Chairman Stevens, and members of the Committee. My name is Brian Staihr, I work as an economist for Embarq, and I appreciate the opportunity to testify before you today.
Embarq is the largest independent wireline telephone company in the country,
serving nearly seven million customers across eighteen states [Fig. 1]. We serve
some of the most rural portions of the country, places like Possum Kingdom, Texas;
Pretty Prairie, Kansas; and Crater Lake, Oregon. And because we serve rural America, we are well aware that this Committee has put tremendous time and effort into
the subject of universal service. Already this year, Senator Stevens has introduced
the USA Act which addresses a number of important issues such as exempting the
Universal Service Fund (USF) from the Antideficiency Act and stabilizing the contribution base while preserving State Universal Service programs. We look forward
to working with Chairman Inouye, Vice Chairman Stevens and all the members of
this Committee going forward as you sort through the complex issues involved in
laying a solid foundation for the next generation of universal service. Getting these
issues right is a matter of vital importance not just to the stakeholders around this
table, but to the economic competitiveness of every rural communityand those in
more populated areas who benefit by connecting to rural America.

74
I want to start out today by highlighting two established facts: First, we know
that the Federal Universal Service Fund has grown dramatically in recent years.
As a result of this growth, the FCC is investigating various ways to control the size
of the Fund, including the use of reverse auctions, which I will talk more about in
a moment.
Second, as the graphs before you illustrate [Fig. 2], we also know this growth has
been driven by the increasing participation of second and third competitive carriers
in the Fund, as opposed to the incumbent carriers that shoulder the core carrierof-last-resort responsibilities.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00078

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

301STAI2.eps

This difference is significant. When a company such as Embarq is a carrier-oflast-resort, that company has an obligation to serve all customers, including the customers in very high-cost areas that no one else wants to serve. Competitive carriers
and wireless carriers do not have these same carrier-of-last-resort obligations, even
when they receive USF dollars.
To illustrate why this difference is significant, Ive included a picture here of a
rural area that Embarq serves called Meadowview, Virginia [Fig. 3]. The different
colors on the left picture represent different population densities, with red and yellow showing the highest densities and green showing low density. As you can see,
the southern portion of Meadowview is actually fairly populous; the northern part
is less populous, very rural, and very high-cost to serve. Embarq, as the carrier-oflast-resort, serves the entirety of Meadowview, the yellow parts and the green parts.

75

In contrast, the picture on the right shows the coverage area of the major wireless
provider in Meadowview. As you can see, wireless coverage essentially stops where
the high-cost areas start. Strange as it may sound, the wireless company has no obligation to serve the high-cost portions of Meadowview, even if it receives USF dollars.
These pictures actually illustrate three related concepts that lie at the heart of
the challenges that Universal Service faces today.
First, before competition, a company like Embarq could serve an area such as
Meadowview and count on the fact that the lower-cost portions would offset the
higher-cost portions. It didnt matter if a company lost money serving the green
areas, because the company also served the yellow areas and, on average, the company could cover its costs.
Second, this changed with the passage of the 1996 Telecom Act. We have seen
competition develop everywhere, but in places like Meadowview the competition is
limited to what we see here: the more densely populated areas. Competitorsboth
wireline and wirelessmost often target the low-cost areas, and avoid the high-cost
areas. As a result, we can no longer count on those lower-cost areas to offset the
highest-cost regions because in many cases weve lost half the customers in the lowcost areas to competition.
Third, and most importantly, the Federal Universal Service Fund has not kept
pace with this competitive reality. When the current USF mechanism evaluates an
area like Meadowview, the system assumes that Embarq can continue to use lowcost areas to offset the higher-cost ones. In fact, the current system assumes that
Embarq can use low-cost areas anywhere in the state of Virginia to offset the cost
of serving the high-cost portions of Meadowview.
In addition, by allowing competitive carriers to receive support while serving only
the parts of a rural study area they choose, the current system creates dysfunctional
incentives that lead to an unfortunate chain reaction:

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00079

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

301STAI3.eps

New entrants gravitate to the town center area and receive support at the same
per-line rate as the carrier of last resort, creating a windfall opportunity;
Drawn by the windfall, multiple competitive providers apply for support in the
same geographic area;
Bereft of its low-cost, offsetting customers, the incumbent carriers per-line costs
go upincreasing the support to all USF recipients in that area, and increasing
the windfall;
The Fund ends up overspending in the town centers and shortchanging the outlying areas where support is most needed.

76
As we look to the future of Universal Service, we need to correct these basic
misassumptions to make the Fund truly compatible with todaysand tomorrows
competitive environment. The way to do that is straightforward: We have to re-examine the geographic area that we use to determine whether support is needed, and
recalculate that support at a much more granular level, so that the town centers
and outlying areas are considered separately, and the support migrates to where it
is truly needed the most. Not only would such an approach eliminate many of those
dysfunctional windfalls, it would be more competitively rational because it would
channel support to the truly rural outlying areas that need it the most, eliminating
those unsustainable cross-subsidies without necessarily increasing the size of the
Fund.
The picture in front of you shows the community of Fort Meade, Florida [Fig. 4].
Each green dot on this picture is a customers location. There is a very clearly identified downtown area which is actually low-cost to serve; then there are outlying
areas where the cost of serving is many times higher. As was the case with
Meadowview, when we see competition in a place like Fort Meade we see it in this
low-cost downtown area. As a result, the outlying areas are the ones that need explicit support from the Fund.

While granular targeting adds a heavy dose of rationality to the USF distribution
process, it is not mutually exclusive to other approaches under consideration, such
as reverse auctions, support for broadband, modifying the identical support rule
or eliminating support for multiple providers altogether. Each of these, and many
other policy decisions associated with universal service, represent important crossroads that will have impact for decades to come. Granular targeting is, however, a
competitively realistic first step for all of those larger decisions that could eliminate
some of the worst abuses and realign the market incentives associated with Universal Service to more closely match the programs original purposeproviding affordable, reliable service where the market would not otherwise deliver it.
Eleven years ago when the Act was passed, true competition hadnt reached any
of the town centers in rural America, and we had neither the tools nor the capability to easily calculate and target support separately for these outlying areas.
Today we have both the capability and the tools. And we have one more thing: The
incentive to do this right, going forward. With that, I will close. Again, thank you
very much for the opportunity to speak with you today, and I look forward to any
questions you may have.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00080

Fmt 6633

Sfmt 6601

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

301STAI4.eps

The CHAIRMAN. I thank you very much, Dr. Staihr.

77
And now, may I call on Mr. Massey?
STATEMENT OF RICHARD N. MASSEY,
EXECUTIVE VICE PRESIDENT, CORPORATE SECRETARY,
AND GENERAL COUNSEL, ALLTEL WIRELESS

Mr. MASSEY. Thank you, Mr. Chairman. And thank you, Mr. Vice
Chairman. And we appreciate very much the interest of you two
particularly, and of the Committee, in Universal Service.
I say that on behalf of Alltel Corporation, and also on behalf of
myself. Im a citizen of one of the more rural states in the country.
And we totally ascribe to the values that both Senator Stevens
billthe USA Act included, and also the Smith-Dorgan-Pryor bill.
We think those move the ball down the field quite considerably,
and we appreciate those efforts.
Alltel is the fifth largest wireless carrier in the country. We serve
about 11 million customers. However, we cover about one and a
half million square miles. Theso, were the largest, in terms of
geography. So, a vast amount of the coverage and our customer
base is rural. We actually have been in the rural businessthe
rural telecom business for 60 years, so we know it pretty well.
What weve learned, spending a great deal of time with our customer base, is that wireless is what they need. Wireless is a very
critical tool that a number of businessmen require to be competitive in this world. If you analyze the industries in a number of
these rural statesI know this is true for Alaska, for an example,
and for Arkansasthe industries in the rural areas are agriculture, mining, timber. These are not desktop businesses, these
are businesses where the employees, the capital, is out in the
world. And what those people tell us is, they want a wireless solution. So, we believe wireless is the future of a lot of the communicationsmaybe not all of it, but a lot of itparticularly with respect
to businesses.
Universal Service has been critical in the development of the
wireless infrastructure in the world. I can tell you, on behalf of
Alltel, there are a number of markets that would not be served but
for Universal Service. So, its very important to us.
Ill give you an example. The Pine Ridge; its Pine Ridge Reservation, South Dakota, is in one of the poorest counties in the United
States. When we found this market, some years ago, it included an
incumbent wireline provider that receives Universal Service funds.
Only 30 percent of the population on this reservation actually used
telephones. We received competitive ETC money, and built the
wireless network there. And today, 80 percent of that population
are wireless consumers.
Senators thats a success story for the Universal Service Fund.
We believe thats what it was intended for. Its to get coverage to
people who cant get coverage otherwise, or they dont choose to get
coverage from another carrier otherwise.
So, broadband deployment, we totally ascribe to the views of all
the Senators and the prior and current committee members who
believe that broadband is the challenge of the future. We believe
it is the interstate highway system of the future. We, however, believe that its not necessarily going to be wireline. We believe the
futurethat broadbands future is in wireless. Today, we have a

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00081

Fmt 6633

Sfmt 6601

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

78
wireless network, as do the Verizon Wireless folks, that provides
data speeds that are comparable to DSL speeds. We believe that,
in many, many marketsin many, many underserved markets,
this sort of technology will be preferred by the consumer.
So, in essence, we believe Universal Service is critical to the community development of wireless in these underserved markets.
Two points. The myriad number of reforms that are here, bewildering thousands of reforms, is very complex stuff. There are two
things that wed like to emphasize here. The first that we would
like for you to make sure is included in any so-called reform by the
Joint Board and the FCC, the first is competitive neutrality. That
was in the Stevens bill. Competitive neutrality is two parts. One
of them is competition. We believe that funding a for-profit monopoly is a bad idea. We think that kind of business went out of style
about 50 years ago. We believe you have to fund some competition
so that subscribersso that customers can get the services they deserve. Neutrality means you dont pick which technology is going
to win in a particular market, you let the customer pick. Thats the
way the Universal Service Fund has worked to date. Customers
pick their carrier. They pick the technology. And we think thats
very important.
Finallyand Im about out of timeaccountability is something
we ask for. We want to be accountable for the funds that you give
us to build out networks in underserved markets. All we ask is
that you impose the same standards on all the carriers uniformly
and fairly. Thats the essence of our proposal on reform.
Thank you very much.
[The prepared statement of Mr. Massey follows:]
PREPARED STATEMENT OF RICHARD N. MASSEY, EXECUTIVE VICE PRESIDENT,
CORPORATE SECRETARY, AND GENERAL COUNSEL, ALLTEL WIRELESS
On behalf of Alltel Corporation, I would like to thank the Committee for inviting
me to speak to you today. Alltel is based in Little Rock, Arkansas, and serves more
than 11 million wireless customers in 35 states. Alltel operates the Nations largest
wireless network in terms of geographic area served, but our customer base is
smaller than those of the larger carriers. This is because we are one of the few
major wireless operators to focus on serving rural and more sparsely populated
areas. We provide leading-edge, digital mobile voice services. We are also rapidly deploying higher-speed, mobile broadband services. Our EVDO based AxcessSM
Broadband service is now available in over 100 communities covering 44 million
peopleincluding numerous high-cost areas where we have been designated as an
ETC. This broadband service offers speeds of 400700 kbpscomparable to the
throughput of many DSL services in the market today.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00082

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

79

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00083

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

301mass1.eps

Alltels roots go back some 60 years as a rural, independent telephone company.


Although we are now exclusively in the wireless businesswe spun off our wireline
local telephone operations last year to the company now known as Windstream
we remain true to our deep commitment to providing the best possible service to
rural Americans. I know there are many other rural-focused wireless carriers across
the country and I acknowledge their great efforts as well.
Mr. Chairman, Mr. Vice Chairman, and members of the Committee, I would like
to commend you for your work in this area. Members of this Committee, past and
present, are largely responsible for the Universal Service provisions enacted in the
Telecommunications Act of 1996 eleven years agoand those provisions have been
a great success in getting affordable telecommunications services, including wireless
services, out to rural communities. The 1996 Act told the FCC and the industry:
lets preserve and advance universal service, and make sure that consumers in
rural, insular, and high-cost areas have access to service that is comparable to services available in urban areas. The Act also said, lets get these services out to people
using a pro-competitive, de-regulatory policy framework and open all telecom markets to competition. These policies are working well. Todays Universal Service system is bringing the most advanced services and technologies, including wireless, to
consumers across Americanot just in metropolitan areas.
Alltel looks forward to working with the entire Committee on Universal Service
reform and I would also like to praise Senator Stevens and the other Members of
this Committee for the introduction of S. 101, the Universal Service for Americans
Act (USA Act). This forward-looking bill sets the right course for Universal Service
policy by reaffirming the fundamental principle of competitive neutrality. Rural consumers will benefit most from a system that promotes Universal Service without
interfering with competition, and without unfairly favoring any class of providers
or technologies over another. We also are enthusiastic about the bills strengthened
eligibility guidelines and auditing provisions. These will increase the programs accountability and will ensure that every dollar of high-cost support is used to maintain and improve communications facilities serving rural consumers. The bill also
wisely broadens the base of Universal Service contributors.
Consumers everywhere increasingly demand mobile, broadband, and other leading-edge telecom and information services. Over the past 5 years, the number of mobile wireless subscribers has grown by 86 percent, from 118 million in June 2001
to 219 million in June 2006. There are now many more wireless phones in service
than wireline. According to a survey conducted by the U.S. Department of Health
and Human Services, over 10 percent of consumers are using wireless as their only
phone service. And among consumers with more than one connection, a substantial
proportion now use wireless as a primary means of communications. Without question, wireless communications is the lifeline of todays consumers. Meanwhile,
wireless broadband service has grown a whopping 2,750 percentfrom about
400,000 lines in 2005 to over 11 million in 2006.

80

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00084

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

301mass2.eps

We are seeing these same trends in rural areas. Rural consumers increasingly
want and need mobile wireless service. Many of you represent rural consumers and
you therefore know that people in rural areas often spend more time than their
urban counterparts on the road, and depend even more heavily on mobile communications, especially since desk jobs are increasingly moving out of rural areas and
into city centers. For example, an entrepreneur may need to reach contacts when
driving from one end of a large county to another for business; a parent may need
access to telecommunications while driving children to and from relatively distant
schools; and a farmer may need access to data on agricultural prices while working
on a remote part of his or her property. Wireless broadband is often the only means
of high-speed access in many high-cost areas and is playing a major role in bridging
the broadband divide. Alltel appreciates the emphasis this Committee places on
the importance of high-speed deployment across rural America.
A critical part of this story is the competitively neutral Universal Service highcost fund program, which, thanks to this Committees efforts, has enabled wireless
carriers to serve the most remote parts of the country. Until just recently, only a
negligible amount of Universal Service funding was going to support the deployment
of wireless service to high-cost areaseven though wireless technology and networks are what consumers in those areas need and want. Of the $25 billion spent
on high-cost Universal Service since 1996, only about $2 billion has gone to wireless
carriers and other competitors. Even today, less than 25 percent of Universal Service high-cost funds go to support the deployment of wireless service, even though
there are now more wireless subscribers. Wireless contributes more than twice the
amount into the Universal Service Fund than it receives out of the fund.

81

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00085

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

301mass3.eps

America is getting a great return on its investment in wireless universal service.


Its true that support for wireless has increased over the past few years. But that
has come with a tremendous expansion of wireless service into rural areas. With
Universal Service support, we are building facilities deep into rural areas, not just
along major highways, and delivering service to consumers where they live and
work. For example, on the Pine Ridge Reservation in South Dakota, the tribe estimated that less than 30 percent of the population had telephone service prior to
Alltels entry into the market as a wireless Universal Service provider. Today more
than 80 percent of the population on the Pine Ridge reservation has access to wireless telephone service. As Senator Thune knows well, the vast majority of these consumers are eligible for and are receiving a discounted Lifeline service of only $1 per
month. This is the true meaning of universal service.

82

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00086

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

301mass4.eps

Similar stories can be told across the country. In rural parts of Arkansas, Louisiana, Maine, North Dakota, West Virginia, and many other states, rural consumers are getting more and better wireless service at increasing broadband speeds
as a direct result of high-cost Universal Service support to wireless companies.
Wireless penetration rates went up from 41 percent in 2001 to 68 percent in 2005
in the most sparsely populated areas with fewer than 100 residents per square mile.
This is a tremendous success story.

83

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00087

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

301mass5.eps

Make no mistakewireless carriers are receiving funds only when we step up and
are held accountable to our commitment to serve the entire geographic area, including outlying areas as well as towns and cities. To obtain ETC (eligible telecommunications carrier) designation and retain that status, we are required to make detailed
annual demonstrations, to the FCC and to most state commissions, that we are
spending the money to build and upgrade cell sites throughout our service areas,
and to maintain and promote top-quality service to consumers in those areas. We
are held accountable for every Universal Service dollar we spend. Alltel added numerous cell sites to its network last year, a significant percentage of which were
the direct result of Universal Service support. Our capital budgeting process considers total funds available, including USF funds, when planning for new cell sites.
Consequently, in each state where Alltel is an ETC, there are several cell sites built
based upon anticipated Universal Service funding. Alltel expends 100 percent of the
USF support on capital and operating expenditures within its ETC areas. And rural
consumers increasingly are depending on wireless ETCs as their carriers of last resort. When we use our USF support to build out new cell sites, we charge the same
amount to everyone who chooses to buy our service; a consumer doesnt have to pay
any more to get mobile service once the network is in place.
Simply put, with wireless high-cost universal service, you get a big bang for your
buck. USF support for wireless in rural areas gives you a great return on your investment. So why do you hear complaints about growing high-cost support for wireless consumers? And why are many parties inundating the FCC and the Joint Board
with proposals that would scale back support for new wireless networks and services in a major way? Alltel urges this Committee to monitor this situation closely
as the Joint Board prepares to make its next set of recommendations. Its true that
the total high-cost fund is growing. But the solutions need to address the real problem. Support for rural wireless is not the problemand anti-competitive proposals
to reduce funding toward wireless consumers are not the answer.
As Verizon correctly noted a few weeks ago to the FCC, the real problem is that
the existing Universal Service program is tailored to support traditional voice-grade
services, while technological changes and increasing competition are transforming
rural consumers telecommunications needs. As a result, the amount of high-cost
funding per lineto wireline as well as wirelessis growing rapidly without efficiently advancing the goals of universal service.
So what is the solution? How can we place reasonable limits on the growth of the
fund, while ensuring that we spend the money wisely and effectively? How can we
do this without harming rural consumers access to competitive wireless and
wireline services comparable to those available in urban areas?
I would like to discuss three policy recommendations that are now under serious
consideration: (1) reverse auctions; (2) placing caps on Fund growth; and (3) tar-

84
geting funds more effectively. Alltel has submitted a proposal (see attached) for reforming how support is distributed from the Universal Service Fund, including a reverse auction aimed at bringing broadband service to unserved and underserved
areas, a per-line cap on Universal Service support for basic voice services, and an
approach to identifying high-cost areas that targets Universal Service support to
those areas and holds carriers accountable for all support received.
1. Reverse Auctions. The amount of Universal Service support could be determined
through a competitive bidding process, rather than through an intrusive regulatory
cost-accounting system. The lowest bid would determine the amount of USF support. Alltel congratulates FCC Chairman Kevin Martin for advancing this innovative idea, which is worthy of further development. As Chairman Inouye and Senator
Stevens have correctly observed in the past, complicated questions arise in connection with auctions for services that are already being provided by existing ETCs,
and there could be serious unanticipated consequences. Alltel believes that it may
be possible to resolve these issues, and ultimately competitively neutral auctions
might be a viable way to set support levels.
Pending the resolution of these broader implementation questions, Alltel has proposed an initial pilot reverse auction program, which would focus on promoting
broadband deployment in the most underserved rural markets. Service providers
using all technologies would bid competitively in a single set of reverse auctions,
and each participating ETC would have to make a commitment to provide substantial broadband service, as well as conventional services, throughout a community
within a specified period of time. The lowest bid would determine the level of perline Universal Service support needed for the auction winner to fulfill this commitment. But other carriers who make the same service commitments would have a
chance to receive some support as well.
The key in this or any USF auction system is to make sure that the competitive
bidding process does not displace competitive service for customers in the marketplace, post-auction. Reverse auctions should be used to set the amount of funding
per line, not to pick a single winner as the exclusive provider of supported universal service. This would give all participating ETCs strong incentives to build facilities and get competitive services out to consumers in rural areas.
Alltel strongly opposes proposals to use reverse auctions to effectively scale down
high-cost funding for one category of Universal Service providerswireless carriers.
For example, consumers would not benefit from the anti-competitive proposal to
hold two separate auctions, the first for wireless only, and the second, presumably
conducted many years later, only for wireline service. This imbalanced type of auction process certainly would reduce support for wireless service in high-cost areas,
by pushing down the level of support per line for a single auction winner, and preventing anyone other than the auction winner from providing supported wireless
services even if it is willing and able to fulfill the obligations of an ETC. The result
would be to dramatically slow the rate of wireless investment in rural areas and
make it harder for rural consumers to access affordable, high-quality mobile service.
But this approach would do nothing to target support to areas where it is most
needed, or to promote deployment of next-generation networks in rural areas. This
Committee should be wary of proposals like this.
2. Caps On Fund Growth. Another proposal under discussion is to place some
kind of caps on the growth of the fund. A cap could be an effective tool in controlling
the growth of the USF, provided that it can be structured in a way that helps rural
consumers. In fact, Alltel has offered a detailed proposal to do just that.
Another version of a fund growth cap has been offered by West Virginia consumer
advocate Billy Jack Gregg, who appeared on the first panel this morning. Under Mr.
Greggs proposal, the total funding disbursed to all eligible telecommunications carriers in a particular geographic areawireless and wirelinewould be allowed to
grow only to the extent that population in the area grows, plus inflation. But the
dollars would be targeted based on the number of consumers who choose to take
service from each ETCthat is, based on the number of lines each ETC serves. If
you serve more customers, then you get more support. If a new carrier comes in and
makes the same ubiquitous service commitment, then it would get a fair share of
the funding as well.
The idea behind both Alltels proposal and Mr. Greggs is, if the country needs
to limit funding growth, then consumers should be the ones to decide where the dollars should flow by deciding what they want to buy, rather than having regulators
make those decisions for them. This way, the competitor that attracts the most consumersby providing the highest quality, most appealing, or lowest cost services
will get the support needed to serve those rural areas.
By contrast, some have suggested that separate caps should be imposed on wireless ETC fund growth and on wireline incumbent fund growth. Like the anti-com-

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00088

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

85
petitive proposal for two separate auctions (wireless and wireline), this proposal
would substantially reduce the amount of funds to support wireless investment in
rural areas, but would fail to satisfy the fundamental principle of competitive neutrality. It might limit the overall growth of the fund, but how would it help rural
consumers? It just continues sending the money where it has always gone, without
doing anything to promote investment and new competitive services in high-cost
areas. Again, we respectfully ask this Committee to be on guard for competitively
biased proposals.
3. Target Funding. A third reform proposal is to target funds more effectively, so
that they would go to carriers that serve consumers who actually live in high-cost
areas, rather than simply giving the funds out based on the same formulas that
have been used for decades. Alltel has offered a detailed proposal to target highcost funding to geographically disaggregated areas, so that funding would flow to
the highest-cost areas in each state, regardless of whether those areas were historically served by large or small incumbents, or by wireless or other competitive carriers. Embarq, to its credit, has offered another, similar proposal, with funding targeted to outlying portions of a study area or wire center, where costs are highest,
rather than to town centers. Re-targeting funding more efficiently would enable the
Fund to support Universal Service goals while also potentially reducing the overall
size of the Fund andmost importantlywithout limiting rural consumers access
to competitive service choices.
Unfortunately, the existing system focuses funding on carriers with high-cost
structures, rather than on consumers in high-cost areas. Some propose to make this
already problematic system even worse, by calculating support for wireless carriers
based on so-called actual costs. This would target the most funds to companies
that spend the most money and punish carriers for providing service more efficiently. It also does nothing to encourage carriers to get services out to consumers.
And it would require a complicated and unnecessary regulatory cost accounting system for competitive wireless carriers. This system doesnt work well today for
wireline incumbents. Why would we want to extend it to wireless competitors?
In conclusion, I would like to thank this Committee once again for its commitment
to policies that simultaneously promote Universal Service and advance competition.
I also appreciate the efforts of the FCC, the Joint Board, and state commissions.
Universal service support is making a real difference in increasing rural consumers
access to wireless services that are vital for health, safety, and economic development. Wireless carriers like Alltel are helping bridge the geographic broadband divide and are enabling rural communities to fully participate in our global economy.
Going forward, Universal Service funds should be targeted and spent more effectivelybut without driving down investments in wireless networks in high-cost
areas. Pro-Universal Service and pro-competitive rules and policies will continue to
bring the benefits of wireless and wireline services to consumers across America.
ALLTEL WIRELESS
February 16, 2007
Commissioner DEBORAH TAYLOR TATE,
Federal Chair, Federal-State Joint Board on Universal Service
Federal Communications Commission
Washington, DC.
Commissioner RAY BAUM,
State Chair, Federal-State Joint Board on Universal Service
Oregon Public Utility Commission
Salem, OR.
RE: HIGH COST UNIVERSAL SERVICE SUPPORT, WC DOCKET NO. 05337
FEDERAL-STATE JOINT BOARD ON UNIVERSAL SERVICE, CC DOCKET NO. 9645
Dear Commissioner Tate and Commissioner Baum:
Consumers in many rural areas rely on high-cost Universal Service support that
carriers use to make available affordable telecommunications services, such as wireless services. Since the entry of competitive eligible telecommunications carriers
(CETCs) into the Universal Service market, rural areas have greatly benefited
from the deployment of basic and advanced wireless universal services. The procompetitive vision of the 1996 Act has become a reality in many rural areas, but
there is more work to be done. As Universal Service reform measures are considered, such as imposing reasonable limitations on the growth of the Universal Service Fund, they must be accomplished without compromising the pro-consumer principle of competitive neutrality. At the same time, Universal Service must continue

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00089

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

86
to evolve to promote the development of new broadband networks and advanced
services.
Alltel submits a set of concrete proposals to advance these goals. We propose the
immediate adoption of a new pilot program of reverse auctions focused on promoting broadband service for consumers in the most underserved, high-cost areas.
Pending development of a broader transformation of the system in the longer term,
we also recommend certain transitional reforms to the existing high-cost support
system that can be implemented immediately, designed to (1) target funding more
effectively to high-cost areas; (2) impose reasonable limits on fund growth; and (3)
ensure greater accountability for the use of funds.
To date the explicit Universal Service funding system has successfully brought
consumers in rural America the benefits of access to robust wireless and wireline
network infrastructure. Our Nations competitively-neutral Universal Service program spurs both wireless and wireline companies to expand their networks and introduce new services for consumers and businesses in rural areas.
At the same time, in rural areas as well as in the rest of the country, technological change and increasing competition are transforming consumers telecommunications needs. Consumers increasingly demand higher-bandwidth services:
across the country, purchases of broadband lines increased by 52 percent from 2005
to 2006, according to recent FCC reports, including an increase from fewer than
400,000 wireless broadband lines in 2005 to over 11 million in 2006. Use of traditional voice-grade wireline telephone lines declined by 3.2 percent over the same
time period. Consumers also increasingly require mobility: mobile wireless service
has grown by 50 percent during the 3 years ending in December 2005, and consumers now use more wireless than wireline lines. Rural consumers have the same
interests in obtaining access to high-speed technologies and mobile services, and are
demonstrating changes in demand that parallel those of consumers across the country. But due to the relatively high costs of deploying wireline and wireless networks
in many rural areas, these services are being deployed less rapidly in rural areas
than elsewhere.
The existing Universal Service system is not well adapted to this changing environment, and a consensus is emerging that the high-cost support rules need reform.
The existing system is designed to support traditional voice-grade wireline servicesfor which demand is shrinkingand does not target funds effectively to promote development of advanced networks in the highest-cost areas. As a result, highcost fund amounts per-line are growing in many areas, without efficiently advancing
the goals of universal service.
The specific and concrete measures we proposebuilding on proposals offered by
Joint Board member Billy Jack Gregg and a range of industry partieswill not only
maintain the availability of existing services in the highest-cost areas, but also will
target funding to promote new broadband services. They will establish greater accountability on the use of support funds and will set reasonable limits to the growth
of the fund. Critically, these proposed measures also remain true to the Commissions core goal and statutory mandate of maintaining a level playing field for facilities-based, intermodal competition to serve rural consumers.
These policy changes will affect CETCs as much as ILECs. Alltel is not offering
these proposals in an intent to benefit or harm any category of providers, but because they will promote the interests of consumers and advance the public interest.
We look forward to working with you on these important matters.
Respectfully submitted,
GENE DEJORDY,
Vice President, Regulatory Affairs.
STEVE R. MOWERY,
Vice President, Public Policy.
MARK RUBIN,
Vice President, Federal Government Affairs.
cc: Joint Board members and staff
Summary of Alltels Universal Service Reform Proposals
Pilot reverse auction system focused on broadband: Use reverse auctions to allocate funds (starting at about $25 million) to bidders that commit to deploy basic and
advanced services, including broadband services (e.g., 400 Mbps) in selected
unserved and underserved markets.
Bidders would offer the lowest amount of funding needed to deploy to specified
proportions of the population in the Zip code within given benchmark dates.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00090

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

87
All ETCsnot just the auction winnercould receive comparable per-line funding if they make the same service commitment.
Reforms to the existing funding system:
To limit fund growth: Allow per-line support in each study area to grow by no
more than the inflation rate.
To target funds more effectively: Disburse high-cost funding to geographically
disaggregated areas, whether served by non-rural carriers or large rural
ILEC holding companies, as well as CETCs:
For purposes of determining funding amounts, consolidate all study areas
served by a single ILEC holding company in each state into a single study
area.
Apply the non-rural funding rules to such study areas if they have more
than 50,000 lines.
Revise the high-cost model forward-looking support mechanism for nonrural carriers (including the consolidated study areas of ILEC holding companies formerly deemed rural) to provide support in the highest-cost wire
centers nationwide, not just in 10 states.
Require all rural ILEC study areas to be disaggregated for purposes of targeting support to the highest-cost portions of such study areas.
To ensure accountability: Broaden the FCCs 2005 accountability and reporting
requirements and apply them to all ETCs, including ILECs as well as FCC-designated CETCs.
Require all ETCs (ILECs as well as CETCs) to document that they are using
their funds to maintain and expand service availability for consumers in highcost areas.
Make USAC, rather than NECA (an RLEC-dominated advocacy group), responsible for collecting and processing cost data and determining support
amounts.
To protect competitive and technological neutrality: Retain the rule that all ETCs
receive the same amount of support per line served.

The CHAIRMAN. I thank you very much, sir.


And now, may I call upon Mr. Tauke?
STATEMENT OF THOMAS J. TAUKE,
EXECUTIVE VICE PRESIDENT, PUBLIC AFFAIRS,
POLICY AND COMMUNICATIONS, VERIZON

Mr. TAUKE. Mr. Chairman, Mr. Vice Chairman, distinguished


members of the Committee, this Committee has shown such great
leadership on this issue in the past. We thank you and commend
you for that. And we are encouraged by your interest in the ongoing challenges with Universal Service.
We often say, in the telecommunications world, that the world
has changed. And, indeed, the world has changed, and its changing
very rapidly. But the Universal Service system and the Universal
Service Fund is stuck in the past. Youve heard a lot today already
about the problems with the Universal Service Fund and the challenges in trying to modernize it for the new age, the new era in
which we live. Id like to offer just a few comments to supplement
the statement that I submitted for the record.
First, I think it is helpful, as we think about these issues, to
think of them in two pieces. One piece is the Universal Service
Fund today and how we fix that Fund. The other piece is, how do
we fund the infrastructure that is needed for the broadband that
we want to deliver to all Americans?
The first piece, the Universal Service Fund, has generally been
focused on maintaining affordable rates for consumers. It is, if you

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00091

Fmt 6633

Sfmt 6601

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

88
will, a supplement to the expense budgets of companies. The challenge with broadband is that you need huge capital expenditures;
help, if you will, with the capital side of the budget. Therefore, attempting to provide the same solution to both, in our view, does not
get us in the right place. So, I encourage you to think of these in
two pieces.
First, on the Universal Service Fund piece, then. Youve heard,
today, that there are a lot of problems with the growth of the Fund.
And, indeed, there are. Id like to share with you just a couple of
thoughts as to why the Fund is growing the way it is.
The first problem that we have with the way the Fund is growing is that we have multiple carriers in many geographic areas.
There are a lot of geographic areas around the country today where
we are subsidizing three, four, fiveand, in some cases, more
wireless carriers, in addition to the wireline carrier. Now, I ask
you, if the public needs to subsidize a carrier to provide service in
a given area, why do we want to just subsidize three, four, or five?
So, the problem is, weve had a proliferation of carriers receiving
subsidies for the same area.
Second problem is, as wireless carriers come in, the Fund explodes because of the nature of the service. If I have a home, for
example, with a wireline carrier in an area that is receiving Universal Service support, and I have, lets say, two lines in that
home, I get support for two lines. If my family has four people who
have four wireless phones, and the wireless carrier applies for a
subsidy, there are four subsidies going into that household. So, the
subsidy doubles from two lines to four lines. The wireless network
is paid on the basis of the cost of the wireline network, even though
the technologies are totally different. And so, the wireless network
is getting twice as much support for that home as the wireline network. This just doesnt make sense. The system needs to be reformed. And this is whats driving the cost that we have in this
high-cost area.
Now, when you look at this problem, and you say you have multiple carriers, particularly wireless, who are receiving this cost that
is defined by wireline, how do you address that issue? We looked
at various ways to do it. Do you want to go through cost proceedings for wireless? How do you choose one of the wireless carriers among all of them? Our view is that the best approach is the
reverse auction concept, so that in areas where you have multiple
carriers, that you look at this reverse auction concept, starting with
the areas where there are multiple wireless carriers, and use that
system to pick which carrier receives the support and also what the
level of support should be.
We think, then the FCC should take a look at how that works
and whether or not that approach should be extended to other
parts of the Universal Service Fund. But there has to be a mechanism to stop the subsidy for multiple carriers and to stop this dependence on the wireline costs to serve wireless.
Second, in the broadband areaand Im almost out of timebut,
in the broadband area, we urge you to take a look at programs like
ConnectKentucky. Kentucky is a tough State to servetough terrain, dispersed population. Through the ConnectKentucky program,
today that state has 94 percent of its homes connected, and expects

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00092

Fmt 6633

Sfmt 6601

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

89
to be close to 100 percent by the end of the year. Theyve done it
by targeting support to areas where there is no broadband service
today, and focusing on getting the capital investment through public-private partnerships and, in some cases, with Federal funds,
into the areas that need the broadband deployment. We think that
approach, of focusing on grants for infrastructure investment, is
the best approach to get quick action in the deployment of
broadband throughout the country.
[The prepared statement of Mr. Tauke follows:]
PREPARED STATEMENT OF THOMAS J. TAUKE, EXECUTIVE VICE PRESIDENT,
PUBLIC AFFAIRS, POLICY AND COMMUNICATIONS, VERIZON
Chairman Inouye and Members of the Committee:
Thank you for inviting Verizon to participate in this hearing on the Universal
Service program for telephony services. Universal Service is a longstanding and appropriate goal of telecommunications policy. However, the means of achieving the
goal of providing affordable telephone service to high-cost areasthe Universal
Service Fundneeds to be reformed.
The world of communicationsdriven by new technologies and competitionhas
changed dramatically and will continue to change. This dynamic process has created
new opportunities for consumers, while challenging all providers in the marketplace
to reinvent themselves. For Verizon, this means investing in new networks, offering
exciting new services to consumers, becoming more customer-focused, and increasing our efficiency in order to compete.
Today the challenges of change are reaching all markets, including those in rural
America. Unlike the days of yesteryear, most consumers in rural America now have
a choice of carriers. But in two-thirds of areas served by rural telcos that receive
Universal Service support, competitive carriers also receive subsidies. In those same
markets, many new providers operate without subsidies.
Unfortunately, the Federal high-cost funding mechanisms intended to ensure that
Universal Service goals are met have not adapted to the changing marketplace. In
fact, these programs are often an impediment to the kind of transformation consumers and the marketplace require. Frankly, the high-cost Universal Service funding system is not working for consumers; its not fair, and we need to work together
to change it.
As competition and technology bring consumers more choices and lower prices,
one would expect that the cost of providing Universal Service would go down. But
its not. Instead, the burden on the consumer to pay the cost of the Universal Service program is going up. The percentage rate of the surcharge on phone bills has
tripled, with more increases on the horizon, and in the past 8 years, high-cost funding has grown from $1.7 billion to $4.1 billiona 142 percent increase.
This increase is driven, in part, by the proliferation of new communications options for consumers. For example, when a family with one wire line buys a wireless
family plan with four handsets, the Universal Service funding provided for that family increases by a factor of five.
Moreover, in many areas we are seeing three, four, even five wireless carriers receiving Universal Service funding. From a public policy perspective, this doesnt
make sense. If the consumer needs to subsidize service in a given area, how many
duplicative infrastructures and carriers should they subsidize? Necessary reforms
must include ways to better target support only to those areas that truly require
subsidies to ensure affordable access.
Another factor that is driving increases in the Fund is that the amount of subsidy
received by wireless carriers is determined by the cost incurred by wireline carriers
to deliver service. To add insult to injury, as wireline telcos lose traditional lines
to wireless, their per-line cost increases, thus driving up the subsidy per customer.
This increased subsidy is then passed on to all providers.
The problem is not just that the Fund is getting bigger. Within the fund, the support for each recipient is also becoming unstable. A telco with cost increases that
are more than the nationwide average can increase its support, while one that
spends less can lose support. This doesnt provide very good incentives for carriers.
Further, in order to keep the telco high-cost funding within its current cap, the
FCC raises the threshold for receiving support. Areas with costs close to the threshold can lose funding entirely as a result. Yet carriers with higher costs are given

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00093

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

90
no incentive to change their behavior. This churn threatens the predictability of
support.
Verizon believes that modernization of the Fund should be guided by the following
principles:
First, funding should be targeted to geographic areas where consumers will be
denied service without universal support.
Second, the Fund should ensure affordable service in high-cost areas, while limiting consumer costs to no more than is required to accomplish that goal.
Third, a new policy should recognize the need to maintain a rural wireline infrastructure even as the number of wireline voice customers declines.
Fourth, a new and fairer system is needed to fund high-cost support.
Reform should start with the way money is collected for the Universal Service
Fund. Verizon supports reform of the pay-in mechanism to the Fund by basing payments on phone numbers. Tying payments to telephone numbers ensures that the
Fund is supported by all voice customers, and it substantially reduces the administrative burden.
We also must reform the way money is paid out of the high-cost fund. Earlier this
month, Verizon filed with the Joint Board a proposal (attached to my written testimony) that would modernize the high-cost funding mechanisms. This proposal
moves us toward achieving the four objectives outlined above.
It meets the needs of rural consumers for high-quality services at an affordable
price. It stabilizes the fund, encourages a competitive and innovative marketplace,
and promotes efficiency so consumers are treated fairly when they pick up the tab
for Universal Service support.
Verizon proposes a reverse auction for the distribution of Universal Service support funds. To ensure an orderly movement to this new system for determining the
payment of Universal Service support, we suggest four steps:
First, we should stabilize funding in each geographic area, by initially capping the
Fund in each area at current levels. This will protect consumers who are paying into
the Fund as we move to a new system. This will also put an end to the instability
and churn of the current fund, making support more predictable.
Second, the FCC should adopt a framework for competitive bidding through a reverse auction. Competitive bidding is the way government generally procures products and services. It allows an agency through a transparent process to select the
most efficient provider and to get the best possible terms. Consumersas users of
rural services and as payers of these servicesbenefit.
Third, this market-based process should begin in areas where there are already
at least two wireless ETCs. The wireless carrier that submits the lowest bid would
enter into a contract, with a specified term, that spells out its obligations. The ILEC
in these areas would continue to receive its existing support, subject to a cap. Once
these auctions have been completed, we suggest that auctions among wireline carriers be held in those few areas where there is a competitive wireline carrier receiving support.
Fourth, after these initial auctions, the FCC should open a new proceeding to review the auction process, and to determine next steps. The FCC might also use the
results of areas where auctions have been held to adjust high-cost support for other
areas.
We believe this approach puts in place a more market-oriented system that will
sustain Universal Service in this competitive marketplace. While todays recipients
argue over costing methods or administrative details of the fund, our proposal focuses every provider in rural areas on the kinds of transformation that produce benefits for consumers: greater efficiency, creative ways of doing business, and new
services.
Let me close with three points on broadband. We all know how important the deployment of new, more capable networks and services is to our future. Verizon is
a leader in that process.
First, we believe that our proposal is the best way to allow the current Universal
Service system to play a constructive role in the deployment of new services. Each
provider in preparing its bid will consider all of the services and revenue sources
in its business plan, regardless of whether they are part of the supported service.
For that reason, the support provided will help the carrier implement all parts of
its business plan. This allows Universal Service to support basic services and encourage broadband deployment in a market-driven way.
Second, recognizing the importance of connecting America to broadband networks,
we believe that we need to approach policies for broadband deployment with great
care, and with an understanding that while broadband is still developing, we are

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00094

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

91
seeing remarkable growth thanks to private investment. Policies that removed regulatory roadblocks have encouraged Verizon and others to invest heavily in new technology.
Third, beyond that, we encourage Congress to review the success of programs to
connect Americans in hard-to-serve areas. Specifically, we call to your attention to
the very successful ConnectKentucky program.
ConnectKentucky pulled the public and private sectors into a partnership which
has already made broadband accessible to 94 percent of Kentucky households.
ConnectKentucky reports that it will increase that number to close to 100 percent
by the end of this year.
The ConnectKentucky program began by compiling an inventory of the current
and planned investment in broadband networks in the state. It then determined if
sufficient demand existed in unserved areas to command private investment. Where
private investment was not likely, the program focused on public-private partnerships and securing public funding from various sources to build broadband facilities.
This program is working because its focused on infrastructure investment. Thats
the key reason why we should not look to the current Universal Service Fund to
solve the broadband issue. The current fund is designed to provide sustained, ongoing support to maintain affordable rates. But maintaining affordable rates is not the
challenge in delivering broadband services to all Americans. Instead, the challenge
in broadband delivery is coming up with the one-time capital investment in infrastructure.
In most places, the private sector is making that one-time capital investment.
Where the market is working, we should let the market continue to meet the needs
of consumers. Where we determine that broadband is not available and the private
sector is not making the needed investment in network facilities, we should target
programs to support infrastructure investment, perhaps through a combination of
loans, tax credits, or grants.
Verizon believes that the process we have proposed will help create a Universal
Service Fund that is sustainable in this new telecommunications marketplace, while
meeting the needs of consumers in high-cost areas, and providing carriers with the
proper incentives to invest and innovate in the communications marketplace.
Verizon looks forward to discussing and working with the Committee on this and
other ideas that further the worthy goals of universal service, particularly in this
time of innovation and opportunity that is being enabled by the communications industry. Thank you.
VERIZON
February 9, 2007
Hon. DEBORAH TAYLOR TATE,
Federal Chair, Federal-State Joint Board on Universal Service
Federal Communications Commission
Washington, DC.
Hon. RAY BAUM,
State Chair, Federal-State Joint Board on Universal Service
Oregon Public Service Commission
Salem, OR.
RE: IN THE MATTER OF FEDERAL-STATE JOINT BOARD ON UNIVERSAL SERVICE,
HIGH COST UNIVERSAL SERVICE SUPPORT, WC DOCKET NO. 05337; IN THE
MATTER OF FEDERAL-STATE JOINT BOARD ON UNIVERSAL SERVICE, CC DOCKET
NO. 9645
Dear Commissioner Tate and Commissioner Baum:
This proceeding is a unique opportunity to put in place meaningful reforms that
will stabilize the Universal Service Fund, create better incentives for companies to
serve rural America in efficient and innovative ways, and lower the cost of access
to communications services for all consumers. The FCC and the Joint Board have
shown constructive leadership on Universal Service reform in order to bring benefits
to consumers and stabilize the fund. It is the right time for these important
changes. More than ever before, consumers of communications services have optionsespecially from new offerings by cable, Voice over IP, and wireless providersand they are taking advantage of them. But at the same time, consumers
are faced with increasing costs as they continue to support a Universal Service system that is growing larger every year.
The need for reform is becoming more urgent as the high-cost fund now surpasses
the $4 billion mark, with approximately $1 billion flowing to competitive eligible

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00095

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

92
telecommunications carriers (CETCs) annually. A solution is needed, and the answer is a system that not only controls the growth of the fund, but provides more
rational incentives to providers and ensures access to important services. Reforms
must also create and sustain an environment that promotes innovation and efficiency gains and makes sure that consumers receive the benefit of these innovations.
For all these reasons, Verizon and Verizon Wireless (hereinafter Verizon) propose that reform should involve the use of auctions or competitive bidding as the
means to better target Universal Service support. This letter proposes the basic
structure for and path to such auctions. Attached is an Appendix that outlines in
greater detail one possible way to design and structure such auctions, although
other approaches and designs may be appropriate and workable.
The reform plan proposed here is a careful and measured approach. It suggests
immediate action to address the most pressing concerns. It proposes implementing
competitive bidding quickly and on a limited basis, and where it can provide the
greatest benefit. It then gives the Joint Board and the Commission the flexibility
to assess the results of these auctions, and to decide whether to extend their use
more widely.
Verizons proposal is as follows:
First, stabilize the Fund by placing a reasonable cap on current support levels
that is designed to control the growth the Fund has experienced in recent years,
introduce better incentives for all ETCs, and prepare for further reform;
Second, establish an administrative framework for competitive bidding, which
would include the auction design;
Third, implement auctions to allocate funding for wireless CETCs. These auctions
would be held in areas that currently support more than one wireless CETC, and
would select a single wireless CETC to receive support. Once these auctions have
been completed, a separate set of auctions should be held for wireline ETCs in areas
where there is currently at least one wireline CETC, to select a single wireline provider of Universal Service for the area.
Fourth, after some reasonable period, the FCC would review the experience
gained with the CETC auctions, and consider developments in technology and rural
markets to determine an appropriate method for extending market-based efficiencies
to additional areas. These methods could include:
A single auction in which both wireline and wireless ETCs would participate,
which would select a single Universal Service provider for each area.
The use of representative bidding, based on statistical analysis of the auction
results, to adjust support for ETCs whose support had not yet been determined
by an auction.
Step One: Stabilize the Fund by Placing a Reasonable Cap on High Cost
Support at Current Levels
As commenters in this docket and many others have observed repeatedly, the
high-cost fund has grown at an alarming pace in recent years and this rate of
growth threatens both the viability and the long-term sustainability of the fund.1
It is also increasing the amounts that consumers must spend on communications
services.
A reasonable cap on the high-cost fund is critical for at least three reasons.
First, the growth in the Fund threatens core Universal Service goals if not contained. The USF contribution factor has risen dramatically in recent years. In 1998,
the contribution factor averaged 3.16 percent and has increased more than threefold since, now standing at 9.7 percent.2 As the Fifth Circuit predicted more than
5 years ago, excess subsidization in some cases may detract from Universal Service
by causing rates unnecessarily to rise, thereby pricing some consumers out of the
market. Alenco Communications v. FCC, 201 F.3d 608, 620 (5th Cir. 2000).
1 The Universal Service Administrative Company (USAC) now projects that in the first quarter of 2007 the high-cost fund will top $4.3 billion. See USAC, HC02High Cost Support Projected by State1Q2007, http://www.usac.org/about/governance/fcc-filings/2007/quarter1
.aspx. This is more than double the size of the Fund just seven years ago. See USAC, Universal
Service Fund FactsHigh Cost Program Data, 19982005 Disbursements by Calendar Year
(2005) (Unaudited), http://www.universalservice.org/about/universal-service/fund-facts/fundfacts-high-cost-program-data.aspx#calendar.
2 See FCC, Industry Analysis & Technology Division, Wireline Competition Bureau, Trends in
Telephone Service, Table Compiled as of April 2005, at Table 19.16 (June 21, 2005); see also
FCC, Proposed First Quarter 2007 Universal Service Contribution Factor, http://hraun
foss.fcc.gov/edocslpublic/attachmatch/DA-06-2506A1.pdf.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00096

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

93
Second, the current high-cost mechanisms do not take into account the benefits
and availability of new competition. Consumers increasingly view cable telephony,
VoIP, and wireless as viable alternatives to wireline phone service. Competition
from these intermodal providers has increased substantially over the last several
years and has brought consumers exciting new services.3 The spread of new intermodal competition in various ways and degrees into all parts of the country has advanced Universal Service goals tremendously. As intermodal competition thrives
and drives down pricessubsidies should be getting smaller or even disappearing
altogether in areas where competitive carriers operate without subsidy. But just the
opposite is happening. Subsidies are increasing even as competition explodes and
rates continue to fall over time.4
Third, a reasonable cap on support at current levels will put in place better incentives for all carriers and allow them to adapt to the new marketplace. The highcost fund in its current form is a product of an earlier time, before competition and
technology transformed the industry. Today, these forces are compelling all providers to become more efficient and more creative, and to develop new services and
new sources of revenue. Yet the current structure of the Fund discourages supported
companies from transforming themselves in a way that advances both their own
long-term interests and those of the customers and communities they serve. Capping support would begin the process of introducing market incentives for innovation and efficiencya process that would subsequently be carried forward through
competitive bidding.
For example, support from the rural high-cost fund is based on a comparison of
each ILECs revenue requirement per line with a nationwide benchmark. This may
have made sense at one time in a less competitive market, but in todays dynamic
market, where the number of traditional telephone lines is shrinking, it is creating
anomalous results and bad incentives:
Under the current rules, as a rural ILEC loses lines, its cost per line increases.
Because CETCs receive the same amount of support per-line as the ILEC, over
time this system also increases the per-line support for each CETCeven
though the CETCs per-line cost is, if anything, falling as it gains customers.
Each rural ILEC can increase its support if its cost per line grows faster than
the national average. This creates an artificial incentive that may bias ILEC
decisionmaking, since the system rewards higher expenditures and penalizes
cost reduction.
The ILEC portion of the high-cost loop fund is capped, but that cap produces
unanticipated effects, creating winners and losers among the ILECs, and a misalignment of incentives. When the total amount of support would otherwise
push the Fund above the cap, USAC raises the nationwide benchmark in order
to ensure that disbursements to rural ILECs do not exceed the cap. This has
the effect of eliminating support for some study areas where per-line costs had
previously been just above the benchmark. The application of the cap thus has
a dramatic impact on the support to those ILECs. Yet ILECs with higher
costswhose spending may have caused the Fund to exceed the caphave no
incentive to change their behavior.
For these reasons, as the first step in the reform process, the Commission should
stabilize the Fund and create better incentives for all ETCs. This can be done by
placing a reasonable cap on the fund, based on current support levels. Support
would be capped for each study area, with two separate caps, one for wireline ETCs
and one for wireless ETCs.
The cap on support for wireline ETCs in each study area would be the total
amount received by all wireline ETCs in that area in a base year, and would include
support from all Federal mechanisms that provide high-cost funding (the high-cost
loop fund (both rural and non-rural), local switching, interstate access support (IAS),
and interstate common line support (ICLS)). If more than one wireline ETC receives
support in a study area, the support amount would be apportioned among them
based on their relative lines.
The cap on support for wireless ETCs in each study area would be the total
amount received by all wireless ETCs in that area in a base year from all the sup3 See Comments of Verizon and Verizon Wireless at 310, WC Docket No. 05337 (filed October 10, 2006) (Comments of Verizon).
4 The Progress & Freedom Foundation, Digital Age Communications Act: Preliminary Proposal
of the Universal Service Working Group, at 910 (Rel. 1.0, Oct. 2005) (footnote omitted) (Although the costs of providing telephone service have fallen significantly over time, [Universal
Service Fund] spending has increased from $15 per household in 1993 to $52 per household in
2003.).

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00097

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

94
port mechanisms listed above. In a study area where there is more than one wireless ETC, the capped support would also be apportioned among them based on their
relative lines.
In order to reflect changes in the overall need for Universal Service in each study
area, each year the total wireline cap and the total wireless cap in the study area
would be adjusted by the percentage change in the number of households in the
area.
The particular structure proposed here, two separate funding limits, applied at
the study area level, will accomplish two important goals: (1) It will end the churn
in supportamong study areas, and between wireline and wireless ETCscaused
by the current rules. As explained above, the current operation of the high-cost loop
fund is producing winners and losers as lines and support amounts change each
year. The more targeted cap described here would minimize those shifts and stabilize wireline support for each study area; (2) By applying separately to wireline
and wireless ETCs, the proposal would curtail what has been the largest source of
growth in the USF in recent yearsnew funding to CETCs.
Step Two: Adopt the Auction Design and Framework
After the cap is in place, the Commission should adopt a framework for the auction process. This framework would include administrative arrangements as well as
the design of the bidding process itself. For auctions to be successful, proper design
is critical. Although the exact details of an auction may be flexible, the following
are the key aspects which are necessary in this context:
Areas for Auction
As part of the framework, the Commission should choose the geographic areas for
which auctions would be held. These areas would then serve as the building blocks
which bidders could, if they choose, package together in the flexible bidding process
described below. Auction areas should be small enough to allow the auctions to target support where it is most needed, but not so small as to create unnecessary complexity. Although other areas of similar size may be appropriate, the most logical
choice among the current alternatives (at least initially) is wire centers. These areas
tend to reflect information about where rural populations are clustered, and thus
distinguish between high and low density areas, since ILEC switches have generally
been located in population clusters, for example in the center of a small town. Although CETCs have different network topologies, they have also tended to locate
their facilities in population clusters for similar reasons, and these areas therefore
tend to be correlated with ILEC wire centers. For this reason, wire centers are a
reasonable choice for the areas to be auctioned.
Package Bids
The Commission should adopt an auction design that allows bidders flexibility to
submit bids for individual wire centers, or bids for packages of wire centers. An auction with this package bidding feature is called a combinatorial auction.
Each bidder will be in the best position, based on its own business plan and market forecasts, to determine whether it is better to bid on individual areas separately,
or in a group or package. By designing the auction this way, the Commission and
the Joint Board would also gain the flexibility to use relatively small, targeted
areas, such as wire centers, as the building blocks for this process. In effect, rather
than deciding itself how these areas should be grouped together, the combinatorial
auction allows the Commission to obtain this information from the market, through
the decisions of the bidders.
By allowing for smaller building blocks such as wire centers, the flexible auction
design would also provide more precise targeting of support, and address concerns
about cherry-picking, without ballooning the fund. At the same time, it would give
CETCs more flexibility to plan their market entry in ways that fit their technologies
and business plans.
Flat Payments to Auction Winners
Auctions for high-cost support should be structured around bids for a flat amount
of support. This approach offers several advantages. First, it eliminates the need to
apportion support among different providers, avoiding controversial issues regarding
whether support should be provided to primary or second lines, wireless handsets,
or on some other basis. It also eliminates one of the main sources of growth in the
Fund in recent years: the addition of multiple handsets by each household.
Each bid can be a flat amount of subsidy for a given area, or package of areas.
This format is simpler and puts the responsibility for estimating demand in a given
area where it belongswith the bidders themselves. ETCs are in a much better position than the auction administrator to know their own revenue expectations and

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00098

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

95
cost structures. In preparing their bids, ETCs will evaluate the competitive landscape and project their own growth should they win the bid to provide supported
services in an auctioned area.
Finally, by providing support in a flat amount, this approach avoids distorting the
incentive each ETC would have to gain or lose a customer. The benefit to any ETC
of gaining a customer would simply be the additional revenue the ETC would obtain
from that customer. Further, the auction gives the Commission, for the first time,
a means to set the flat support at the amount that is just sufficient to make an
ETC willing to undertake the burden of the Universal Service responsibility. Taken
together, these features ensure that the proposed framework would not distort competition at the margin among ETCs in an area and would not prevent competition
from occurring in an area that would otherwise have supported it.
Auction Reserves
Any auction for Universal Service support should include a reserve amount, which
is the maximum bid that would be accepted. Reserves are commonly used in auctions to limit the range of possible outcomes. In the Universal Service context, the
reserve ensures that the support determined by the auction is no greater than the
amount of support provided prior to the auction.
The reserve reflects the limit of what the auction administrator would be willing
to pay. By selecting the most efficient provider, and identifying the support amount
that provider is willing to accept, the auction offers the best opportunity to obtain
Universal Service on terms most advantageous to the public. However, if no bid
lower than the current support amount is submitted, the administrator is better off
reverting to the existing support arrangement, which would continue in an auctioned area where the reserve is not met.
The auction design included here suggests two reserves that would each have to
be satisfied: one that applies at the study area level, and a second reserve that applies at the wire center level. The aggregate reserve at the study area level would
be the capped amount established at the beginning of the process. The wire center
reserve should be based on a pro-rata distribution of the study area support to each
wire center, but with some additional amount added to allow for the auction results
to direct more support to higher cost wire centers, and less to lower cost ones. This
means that the sum of the individual wire center reserves in a study area would
be greater than the aggregate reserve for the study area as a whole. However, the
separate imposition of the study area reserve would ensure that the auction cannot
result in an increase in support for any study area.
Step Three: Auctions for Wireless and Wireline CETCs
It makes sense for the Commission and the Joint Board to start, as an initial step,
with auctions for wireless CETCs in areas in which multiple wireless CETCs currently operate and receive support. This would be followed by a parallel set of auctions for wireline ETCs, in areas where at least one wireline CETC has been designated.
Wireless CETCs operate on fundamentally different cost structures than ILECs
a fact that has long made the Commissions portability rules, which tie CETC support to the ILECs per-line costs, a primary target for reform. Starting the competitive bidding process with wireless CETCs would immediately help to connect wireless CETC subsidies with the actual cost of providing wireless services, as wireless
CETCs bid against each other for support in those areas eligible for auction. A wireless CETC auction will ensure that affordable wireless service is available in highcost areas, and that such service is provided by the most efficient wireless provider.
Using an auction to select a single wireless CETC in each area is an important
step toward rationalizing distributions from the fund. Support to CETCs (primarily
wireless carriers) has caused substantial growth in the Fund over the last few years.
In 1999, wireless carriers received approximately $500,000 in high-cost support.5 By
2002, wireless CETC support had increased to approximately $45 million. Id. In
2005, wireless CETCs received more than $600 million in high-cost subsidies and
through May of last year, that number increased to more than $800 million. Id. At
this rate, CETCs will soon account for approximately 25 percent (if not more) of all
high-cost subsidies. While in many areas a wireless CETC may ultimately prove to
be the most efficient provider of universal service, funneling more and more support
to fund duplicative networks in high-cost areas should not continue. With wireless
5 See USAC, Distribution of High Cost Support Between Wireless and Wireline CETCs,
http://www.universalservice.org/lres/documents/about/pdf/fundfacts-High-Cost-Support-Between-CETCs-199 8-2006.pdf.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00099

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

96
carriers and their customers now paying a significant share of the Federal USF,6
wireless consumers will be harmed by continual increases in USF assessments. The
public interest will be served by stabilizing the Universal Service Fund and directing wireless subsidies to the most efficient providers through the use of competitive
bidding.
The Commission should also allow for a reasonable transition for wireless CETCs
that are receiving support today, but do not receive support after the auction. The
ILEC, and any wireline CETC in that area, would continue to receive support on
the basis of the capping mechanism established in Step 1.
Once the wireless CETC auctions have been completed, the Commission should
also nominate for auction any area where there is at least one wireline CETC. In
these auctions ILECs and wireline CETCs would participate, and each auction
would select a single wireline provider of Universal Service for the area. The reserve
for this auction would be the total amount of support received by wireline ETCs in
the area prior to the auction. These auctions would be held in a relatively limited
number of areas, since wireline ETCs are designated in about 90 study areas today.
Step Four: the Commission and the Joint Board Review Auction
Experiences and Decide Next Steps
After some reasonable period, the FCC should initiate a review of its experience
with the wireless and wireline CETC auctions. The Commission would consider the
development of markets in rural areas and changes in technology and determine
next steps. Options would include:
1. Conducting general auctions. The Commission could decide to move forward
with general auctions in which both wireline and wireless ETCs would participate.
Such an auction would be held in each high-cost area where there is at least one
CETC, and would select a single Universal Service provider for the area to receive
the support determined by its bid.
2. Using representative bidding. The Commission could use the results of auctions,
where they have been held, to adjust the support of ETCs whose support has not
yet been established by an auction. This use of representative auctions is an established practice in other applications.7 Once it has assembled a representative sample of results from the areas where bidding has been completed, the FCC could commission an econometric study that would relate the auction results to the characteristics of a high-cost area, such as size and density. This econometric model would
estimate the likely results of an auction in an area with given characteristics.
Results from wireless auctions could be extended to wireless CETCs operating in
areas where auctions had not yet been completed. Results from wireline auctions
could be applied to wireline ETCs whose support had not yet been set by auction.
The support amount for these ETCs would then be set at the lower of the capped
support amount or the amount estimated from the auction results. If an ILEC believes that the estimated support should not be implemented in a given area, it
would have the option of nominating the area for an auction.
*
*
*
*
*
*
*
In its present form, Universal Service funding provides companies with the wrong
incentives, discourages innovation, and has increased the amounts consumers pay
for communications services. The approach outlined here will help remedy these
problems and transform the Fund into an efficient, market-oriented system that advances the core Universal Service objectives.
Sincerely,
KATHLEEN GRILLO,
Vice PresidentFederal Regulatory.
ATTACHMENT
cc: Chairman Kevin J. Martin
Commissioner Jonathan Adelstein
Commissioner Michael J. Copps
Commissioner Robert M. McDowell
Hon. Lisa Polak Edgar
Hon. Larry S. Landis
Hon. John D. Burke
Hon. Billy Jack Gregg
Daniel Gonzalez
6 See
7 See

VerDate Nov 24 2008

14:09 Mar 21, 2012

Michelle Carey
Ian Dillner
Scott Bergmann
Scott Deutchman
John Hunter
Thomas Navin
Donald Stockdale
Amy Bender
Jeremy Marcus

Alltel Ex Parte Presentation, CC Docket No. 9645 (Oct. 20, 2006) at Attach. at 12.
Comments of Verizon and Verizon Wireless at 2728.

Jkt 073387

PO 00000

Frm 00100

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

97
Vickie Robinson
Ted Burmeister
Katie King
Gary Seigel
Phil Nyegaard
Jacob Williams
Jennifer A. Richardson
Peter Bluhm
Peter A. Pescosolido
Joel Shifman

Jeff Pursley
Lori Kenyon
Aram Shumavon
Eric Seguin
Brad Ramsay
David Dowds
Michael H. Lee
Philip McClelland
Denise Parrish
APPENDIX

Modernizing Universal ServiceA Design for Competitive Bidding


This appendix illustrates one way the Joint Board and the FCC could implement
a competitive bidding process for Universal Service obligations.
1. Summary
The auction design outlined in this appendix would introduce a more efficient
framework for the distribution of support to Universal Service providers in high-cost
areas. This could be done in a series of steps:
First, immediate measures would be taken to stabilize the fund, and to introduce
better incentives for all ETCs, by capping support based on current levels.
Second, the FCC would adopt a framework for competitive bidding, including administrative arrangements and the design of the bidding process itself.
Third, to initiate the use of competitive bidding, the Commission would prompt
auctions in high-cost areas where there are multiple wireless CETCs. These auctions would select a single wireless provider of Universal Service for each area. The
incumbent local exchange companies in those areas would continue to receive support based on the capping mechanism. Once the wireless CETC auctions had been
completed, the FCC would also nominate any area where there is at least one
wireline CETC. These auctions would select a single wireline provider of Universal
Service for each of those areas.
Fourth, after some reasonable period, the FCC would review the experience it had
gained with the CETC auctions, and consider developments in technology and rural
markets to determine an appropriate method for extending market-based efficiencies
to additional areas. These methods could include:
A single auction in which both wireline and wireless ETCs would participate,
which would select a single Universal Service provider for each area.
The use of representative bidding, based on statistical analysis of the auction
results, to adjust support for ETCs whose support had not yet been determined
by an auction.
2. Stabilize the Fund
The FCC should start by taking immediate steps to stabilize the fund, bring fund
growth under control, and put in place incentives for all ETCs to adapt to changes
in the market and become more efficient. This would establish a starting point for
the implementation of competitive bidding.
Support would be capped for each study area. There would be two separate caps
in each study area, one for wireline ETCs and one for wireless ETCs.
Cap for wireline ETCs. The cap on support for wireline ETCs would be the total
amount received by all wireline ETCs in the study area in a base year (which
could be the most recent twelve-month period for which data are available when
an order becomes effective). The cap would include receipts from all programs
for high-cost areas (the high-cost loop fund (rural and non-rural), local switching, interstate access support (IAS), and interstate common line support
(ICLS)).1
1 For ILECs, once the cap described here has been applied, it would replace the calculation
that is done today to determine support amounts from each of the existing funds. The exception
would be the calculation for rate-of-return ILECs of the support amounts for local switching and
ICLS, which would be calculated as they are today. High cost subsidies in each rate-of-return
study area would then be adjusted to bring the total amount of support within the study area
cap. The current cap on the ILEC portion of the high-cost fund would no longer be applied. For
price cap ILEC study areas, the total amount of wireline support in each area should simply
Continued

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00101

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

98
If there is more than one wireline ETC in the study area, the capped support
amount would be apportioned among them on the basis of their relative lines.
The current cap on the ILEC portion of the high-cost fund is producing winners and losers as lines and support amounts change each year. The mechanism described here would minimize those shifts and stabilize wireline support for each study area.
Cap for wireless ETCs. The cap on support for wireless ETCs would be the total
amount received by all wireless ETCs in the study area in a base year (which
could be the most recent twelve-month period for which data are available when
an order becomes effective). The cap would include support from all programs
for high-cost areas (the high-cost loop fund (rural and non-rural), local switching, interstate access support (IAS), and Interstate Common line support
(ICLS)).2
If there is more than one wireless ETC in the study area, the capped support
amount would be apportioned among them on the basis of their relative lines.
Increased support for wireless ETCs represents a large proportion of the
growth in the Federal mechanisms in recent years. The cap would stabilize
the Fund and provide a starting point for the wireless ETC auctions.
Adjustment of the caps. Each year, the total wireline cap and the total wireless
cap in each study area would be adjusted by the percentage change in the number of households in the study area. This would allow the cap to reflect changes
in the overall need for Universal Service in the area. However, there would be
no adjustment for the total number of lines or handsets in the area. The current
rural growth factor (which has been negative in some recent years) would be
eliminated.
3. Adopt the Framework
Before any auction takes place, the FCC should adopt a framework for the auction
process.
a. Areas for Bidding
The FCC would first designate the geographic areas that would be used for bidding. Areas should be small enough to allow support to be targeted where it is most
needed, but not so small as to create unnecessary complexity. They should incorporate information about where rural populations are clustered, so as to distinguish
between high and low density areas.
Geographic units such as census block groups or counties are possibilities, but
these areas often cut across geographic barriers, such as mountains and rivers, and
ignore clustering of customers that would be relevant to any prospective provider
of universal service. The arrangement of ILEC wire centers, however, contains useful information about the geography of each area and the location of customers,
since ILEC switches have generally been located in population clusters (in the centers of small towns). CETCs, while they have different network topologies, have also
tended to locate their facilities in population clusters for similar reasons; they have
put their facilities where the customers are.
The use of ILEC wire center areas represents a reasonable balance among these
considerations. If some other geographic unit of similar size is readily available, and
meets the requirements discussed here, then the Commission may consider that
unit in place of wire centers. Once a geographic unit has been selected, steps should
be taken to ensure that all potential participants in an auction would have ready
access to data delineating the boundaries of those areas. An auction design that allows for package bids (as discussed below) makes it possible to use areas that are
smaller than a study area.
b. The Reserve or Maximum Bid
The Commission would also establish a maximum bid, or reserve, for each wire
center. Reserve amounts are widely used in competitive bidding processes to limit
the range of possible outcomes. In this case, the reserve amount would be set at
the level of the support provided immediately prior to the auction. In this design,
two reserves would be enforced: the first at the study area level, and the second
at the wire center level.
be capped, and if there are wireline CETCs in the area the support would be apportioned among
the wireline ETCs on the basis of their relative lines.
2 For wireless ETCs, none of the existing funds is capped today. The total amount of funding
to wireless CETCs in each area should simply be capped, and the apportionment among wireless
CETCs on the basis of their relative lines would replace the existing fund calculations.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00102

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

99
The aggregate reserve. For the wireless auction, the aggregate reserve for each
study area would be the total amount of support provided to all wireless ETCs in
the study area prior to the auction. For the wireline auction, the aggregate reserve
for each study area would be the total amount of support provided to all wireline
ETCs in the study area prior to the auction.
The wire center reserve. In order to allow competitive bidding to proceed at the
wire center level, it would be necessary to develop a reserve amount for each wire
center. This would be done by disaggregating the existing support at the study area
level in the following way:
First, the aggregate reserve in the study area would be divided by the total
lines of all wireless (wireline) ETCs to derive an average per-line support
amount.
Second, the aggregate study area reserve would be disaggregated to each wire
center on a pro-rata basis by multiplying the number of wireless (wireline) ETC
lines in each wire center by the average per-line support amount.
Finally, each wire center amount would be multiplied by a constant greater
than one to arrive at the wire center reserve amount.
This approach allows a reserve to be developed for each wire center, but avoids
the need for the Commission to develop detailed cost estimates by wire center.3 Because each wire center reserve is greater than its pro-rata share of the current level
of support in a study area, it also provides room for the bidding process to provide
more support to higher cost wire centers, and less support to lower cost ones. However, this also means that the sum of the individual wire center reserves will be
greater than the aggregate reserve at the study area level. The application of the
aggregate reserve ensures that the bidding process cannot result in an increase in
support for the study area as a whole.
c. Qualification Process
Qualified bidders that would be eligible to participate in the bidding process
would be providers who have been designated as ETCs in the area. This is consistent with Section 214(e), which requires a carrier to be an ETC in order to be
eligible for support.
d. Obligation of the Auction Winner
In any competitive bidding process, the ETCs would be bidding for the obligation
to serve as the provider of Universal Service in a high-cost area, in return for which
it would receive financial support equal to the amount of its bid.4 The Commission,
in cooperation with the states, would develop a statement that would define the
winning bidders obligations. This would, in effect, serve as a request for quote (or
RFQ).
In return for the Universal Service support, the winning bidder would be required
to offer service in the entire area, and to meet any other terms of the RFQ. If a
wireless CETC bids for an area and loses, then that CETC would no longer have
an obligation to serve that area.
e. Schedule and Organization of the Bidding
In this design, competitive bidding would not take place simultaneously in all
areas. Instead, bidding would be introduced gradually through a series of transitional steps.
The Commission would establish a regular schedule of events leading up to an
auction. This would include nomination of areas for bidding, registration of bidders,
posting of deposits, and the bidding process itself (this series of events is referred
to here as a bidding cycle). This flexible framework would allow the Commission
to manage the transition to competitive bidding in reasonable steps, and, at the
same time, provide ETCs themselves with the opportunity to decide when an area
is ready for competitive bidding.
A bidding cycle would be held twice each year. The first bidding cycle would
begin 6 months after the adoption of an order establishing the plan.
3 The Commission does not need to engage in detailed cost analysis in order to establish reserves. In fact, part of the reason to use competitive bidding is to reduce reliance on traditional
measures of cost. However, auction results might be improved if some simple indicator could
be developed, perhaps based on the size or density of the wire center, to differentiate between
higher and lower cost wire centers. Support from the non-rural high-cost fund is already
disaggregated to the wire center level. There is also a process in place for ILECs to develop and
submit proposals to disaggregate study areas for USF purposes, and where such plans have
been approved, they could be used to calculate a reserve at the wire center level.
4 Some of the Universal Service mechanisms, such as Lifeline, Link-Up, schools and libraries,
and rural health care, are not related to high-cost subsidies, and would not be determined
through the competitive bidding process outlined here.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00103

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

100
In any cycle, a wireless CETC would be able to nominate for bidding any area
for which it is qualified, and where there is at least one other wireless CETC,
except in areas where an auction had already been held and the term of the
contract resulting from that auction had not yet expired. A wireline ETC would
be able to nominate an area where there is at least one wireline CETC for a
wireline auction, except in areas where an auction had already been held and
the term of the contract resulting from that auction had not yet expired.
At certain points in the transition process, the Commission would, on its own
motion, nominate areas that meet certain criteria. For example, as discussed in
Section 4, it would nominate areas with more than one wireless CETC to begin
the wireless CETC auctions.
Dates would be established for the events in each cycle. For example, if a wireless CETC wished to nominate an area for bidding in the first half of a given
year, it might be required to file its nomination by February 1 of that year.
Once an area has been nominated, a second window would be established for
ETCs to register to bid in areas that had been nominated, and to nominate additional areas. This would prevent an ETC from gaining a first-mover advantage by nominating an area, would ensure that all ETCs interested in a given
area are able to participate, and ensure that all areas related to those initially
nominated can be included in the bidding process.
The Commission would set a firm date for bidding to begin. As described in Section 6 below, bidding would be dynamic, which is to say it would involve multiple rounds.
By grouping all of the bidding processes for each six-month period together, this
framework would simplify administration. And, by announcing a clear schedule
of events in advance, the framework would also make it easier for ETCs to plan
their participation in the bidding process.
4. Auctions for Wireless and Wireline CETCs
To initiate the use of auctions for universal service, the Commission could first
prompt competitive bidding among wireless CETCs.
In each area where there is more than one wireless CETC, an auction would select one winner to be the wireless provider of Universal Service in that area. Any
area that had not previously been nominated by a wireless CETC, and where more
than one wireless CETC is already certified, could be nominated by the FCC on its
own motion.5 Wireless CETCs would bid for a flat amount of support in each area.
The design of the bidding process is discussed in Section 6.
Once a wireless winner is selected, that provider would receive the support
amount contained in its bid. The ILEC, and any other wireline ETC in the same
area, would continue to receive support under the cap mechanism described in Section 1.
The FCC could publish results of all auctions on a website, where that information would be available for use by any bidder in formulating its bid in subsequent
auctions.
Once the wireless CETC auctions have been completed, the Commission should
nominate for auction any area where at least one wireline CETC has been designated. In these auctions, both the ILEC and any wireline CETC would participate,
and the auction would select a single wireline provider of Universal Service for the
area.
5. FCC Reviews Auction Experience, Decides Next Steps
After a reasonable period, the FCC could then review its experience with the wireless and wireline Universal Service auctions.
The Commission would consider this experience, the development of markets in
rural areas, changes in technology, and the acceptance of substitutes by customers
of different services.
Based on this experience, the FCC would then determine whether it should nominate additional areas for auction.
A general auction. The Commission could prompt a general auction in any area
where there is a CETC. Both wireline and wireless ETCs would participate. The
general auction would select a single ETC to be the Universal Service provider
for the high-cost area and to receive the support determined by its bid. The auction design described here is intended to be suitable for a general auction; the
5 The Commission could decide either to prompt bidding on all such areas in one bidding cycle,
or could decide that it would be more convenient to spread the auctions out over time.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00104

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

101
FCC could determine whether any adjustments would be appropriate, based on
the experience gained with previous auctions. The reserve for this auction could
be the sum of the wireline and wireless support amounts provided on the date
of the general auction.
Representative bidding. As part of its review, the Commission should also consider whether to use the results of auctions, where they have been held, to adjust the support of ETCs receiving support not yet established by an auction.
Once it has assembled a representative sample of results from the areas where
bidding has been completed, the FCC should either perform or commission an
econometric study that would relate these results to the characteristics of the
areas, such as size and density. This econometric model could then be used to
estimate the likely results of an auction in an area with given characteristics.
Estimates based on the wireless auctions, or on general auctions, could be used
to adjust the support of a wireless ETC in an area where a wireless ETC auction had not yet been completed, (either because the area had not been nominated, or because an auction in the area had failed).
Estimates based on the wireline auctions, or on general auctions, could be used
to adjust the support of wireline ETCs whose support had not yet been set by
an auction.
The support would be the lower of the capped support amount or the amount indicated by the econometric study.6 If an ETC does not believe that the estimate produced by the econometric study should be applied to a given area, then it would
have the option of nominating that area for bidding.
6. Design of the Competitive Bidding Process
The design outlined here is called a clock-proxy auction. The bidding process
would be a hybrid of two designs that combines the advantages of each. The first
phase is a clock auction. The second phase is a proxy auction. This design draws
on the latest work of auction experts in this area (including the Commissions own).
A similar design has recently been adopted by Ofcom for a major spectrum auction
in the United Kingdom.
a. The Clock-Proxy Hybrid
The last few years have seen significant advances in auction design theory.7 One
of these advances has been the development of a hybrid of two types of auction designs, a clock auction and a proxy auction. This hybrid is called a clock-proxy
auction.
The first phase of this design would be a clock auction. A clock auction is a dynamic, multiple round process in which the auctioneer announces prices and bidders
respond with quantities desired at the announced prices. It is called a clock auction
because the rounds of bidding are conducted at regular intervals. This design allows
the auction itself to generate information useful to the bidders. By observing the results of the early rounds, each bidder gains knowledge of the value of each area and
how the areas are related to one another. In this respect, the clock phase of this
design is similar to the spectrum auctions. Importantly, a clock auction also limits
the opportunities for bidders to engage in strategic behavior compared with a more
conventional multiple-round auction in which the bidders themselves formulate the
bids. In each round, a bidder can only answer a yes-or-no question for each area
or package of areas: will the bidder be willing to become the Universal Service provider at the support amount called out by the auctioneer? This kind of design thus
makes it difficult, for example, for a bidder to use the amount of its bid to signal
other bidders.
The second phase of this design would be a proxy auction, which is based on
the results of the clock phase. The proxy phase is necessary to make the results
from the clock phase more efficient. It provides the opportunity for bidders to create
combinations of prices that would not have occurred in the clock phase. This is
6 As Verizon and Verizon Wireless noted in their comments, this approach has been used to
extend auction results in other settings, such as the pricing of timber cutting rights in Canada.
Comments of Verizon and Verizon Wireless at 2728, WC Docket No. 05337 (filed October 10,
2006).
7 For an overview of modern auction theory, see Paul Milgrom (2004), Putting Auction Theory
to Work, Cambridge: Cambridge University Press. For essays on various aspects of combinatorial
auctions, see Peter Cramton, Yoav Shoham, and Richard Steinberg (2006), Combinatorial Auctions, Cambridge, MA: MIT Press. A discussion of the clock-proxy design is provided in Lawrence M. Ausubel, Peter Cramton, and Paul Milgrom, The Clock-Proxy Auction: A Practical
Combinatorial Auction Design, which appears as Chapter 5 in Cramton, Shoham, and Steinberg.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00105

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

102
called the proxy stage because the bidding activity is conducted by a proxy agent
(a computer program) following strict rules in order to limit the possibility of strategic behavior by the bidder itself.
b. Advantages of the Clock-Proxy Hybrid Design
Flexible bidding for individual areas, or packages of areas. This design allows the
bidders to place bids on different areas in a very flexible way. A bidder could submit
bids on a specific area or areas. The same bidder could also submit a package bid
on a group of areas, if the bidder found them to be related to one another (for example, if the bidder could serve the package more efficiently than the individual
areas separately). This type of bidding process is called a combinatorial auction.
A design which permits the flexibility of package bidding makes the choice of the
area to be auctioned less critical. It would allow the Commission to design the auction around smaller geographic units (such as the wire center areas discussed here)
without unduly complicating the bidding process. Rather than having the Commission make decisions about how areas should be grouped together, this approach allows the Commission to elicit information from the bidders about how the areas
should be grouped. This design would achieve more accurate targeting of Universal
Service support, and address cherry-picking concerns. These advantages would be
gained without inflating the fund, and without giving up the economies of serving
larger areas in cases where those are important.
Allowing for different relationships among areas. The auction design outlined here
is designed to perform wellin terms of efficiency, and minimizing the need for supportregardless of whether different bidders view a given set of areas as independent, substitutes, or complements. This is important because in bidding for Universal Service support, all three of these are possible:
Areas are independent if a bidders willingness to bid for hypothetical area A
is not affected by the outcome of the bidding for any other area. For example,
a small ILEC that serves a single wire center may care only about that area.
Two areas are substitutes if a bidder wishes to win either area A or area B,
but not both. This could be the case for a wireless carrier that wants to enter
one new market, and is considering A and B as possible alternatives. If in the
early rounds of bidding this carrier encounters strong competition for A, it may
shift its attention to B in later rounds. This kind of behavior has occurred in
the spectrum auctions.
Two areas are complements if a bidder sees some synergies in serving the two
areas together, so that it would be willing to accept less support in area A if
it also wins area B. For example, a mid-size ILEC that serves several wire centers in a state may view them as complements. In this case, strong competition
for A may make this carrier less willing to bid for B.
Some earlier proposals for competitive bidding of Universal Service have essentially treated high-cost areas as independent.8 For that reason, they do not make
any provision for either substitutes or complements. The multiple-round design used
in the spectrum auctions performs well when areas are substitutes, but not as well
when they are complements. As explained in more detail below, the clock-proxy auction design will perform well regardless of whether different bidders view a given
set of areas as independent, substitutes, or complements.
Minimizing strategic behavior. The design outlined here also minimizes the possibility of strategic behavior, such as collusion among the bidders, or an attempt by
one bidder to conceal its interest in particular areas by holding back until the late
rounds of an auction. This is particularly important in the context of bidding for
universal service, where the number of bidders for any given area is likely to be
small. Because this design encourages each party to bid straightforwardly based on
relevant business factors, such as its expected costs and revenues, it would improve
the transparency of the process, and the efficiency of the outcome.
Single Winner-Flat Amount of Subsidy. This design allows for a single winner.
Thus, there would be no need to attempt the difficult task of apportioning support
8 For example, neither Milgrom (Paul Milgrom, Procuring Universal Service: Putting Auction
Theory to Work, Lecture at the Royal Swedish Academy of Sciences, December 9, 1996) nor
Weller (Dennis Weller, Auctions for Universal Service Obligations, Telecommunications Policy,
Vol. 23, 1999, pp. 645674) allowed for package bidding; instead they proposed a separate auction for each area. Since these designs were also single-round, sealed-bid auctions, they did not
allow bidders to shift their attention from one area to another based on results in earlier rounds.
The only provision for complementarity was a limited opportunity for a bidder to withdraw if
it wins area A but loses some other area it sees as related. Because the design proposed here
deals directly with package bidding, and also allows for multiple rounds, there is no need for
such a withdrawal provision.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00106

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

103
amounts among different providers. This would avoid many contentious issues that
have arisen in the past, such as whether to support primary lines, additional lines,
multiple handsets, and so on. It would also make for a simpler bidding process.
Each bidder would bid a flat dollar amount of subsidythe total amount the ETC
would accept in order to take on the Universal Service obligation for a given highcost area. Each bidder would base its bid on its own business plan, which would
include the bidders own assessment of many factorsincluding the demand quantities (of lines, handsets, etc.) it would expect to serve within each area.
c. Clock Phase
As discussed above, in the first phase of the auction (the clock phase), the bidding would proceed in a series of discrete rounds. Instead of having the bidders submit support amounts, the auctioneer calls out a support amount for each area in
each round. Each bidder then indicates which areas it would be willing to serve as
the Universal Service provider at the specified support amount. The clock phase
would proceed as follows:
The support amount called out by the auctioneer in each round is a flat amount
per year. It is constant each year for the duration of the contract. In the first
round of the clock phase, the auctioneer calls out the reserve price in each wire
center.
In each round of the clock phase, each bidder may submit a bid on a package
that includes any area or combination of areas it chooses. Since the support
amounts are being announced by the auctioneer, the package bid is simply a
list of the areas the bidder would be willing to serve for the amounts called out
in that round. Each bid is also exclusive in the sense that at the end of the
clock phase the auctioneer can accept only one bid for each area, and one bid
from each bidder. All bids remain in effect for the entire duration of the auction
and cannot be withdrawn (even after bidding has closed). At the end of the bidding process, the auctioneer may go back and accept any bid from a previous
round. This means that a bidder must carefully consider what it bids in every
round, because every bid is a binding offer that the bidder might be called upon
to honor.
At the end of each round, the auctioneer determines how many bids have been
submitted for each area. The objective of the auctioneer is to select a single bidder for each area. Therefore, in an area where more than one bid has been received, there is excess supply. In areas where no bids have been received there
is excess demand. In areas where there is excess supply (more than one bidder)
the auctioneer reduces the support amount called out in the next round by a
set amount.9
The auction is held over the Internet, using a software program to administer
the bidding.10 The program includes admission control to ensure that only
qualified entities submit bids. The program also checks to see that bids meet
the rules, and prompts the bidder to resubmit a bid if it does not. The rounds
occur at some set interval, perhaps every 2 hours.
The program will accept only bids that meet the wire center reserve. It also
checks after each round to see that the aggregate reserve is met at the study
area level, and provides that information to the bidders prior to the next round.
This aggregate reserve check can only be done after a round is completed, so
within a round each bidder does not know if the bids being submitted, taken
together, will satisfy the rules. In some cases, not all wire centers in a study
area will have been nominated for bidding. In this event, in order to apply the
aggregate study area reserve, the auctioneer would include the areas that were
not part of the auction in the calculation as if they had received bids at their
wire center reserve amounts.
Each bidder would be subject to an activity rule, which would require it to
bid actively in every round in order to maintain eligibility to bid in subsequent
rounds. This rule, which has been used in the spectrum auctions, prevents a
9 The decrement by which the bid is reduced each round is an element of the auction design.
A large, or coarse, bid decrement will make the auction go faster, but may jump over the correct
support amount. To address this issue, a device called intra-round bidding may be used to obtain finer information from the bidders. Rather than simply drop out of the bidding for an area
when the support amount falls below the level it would accept, a bidder could indicate willingness to accept a level of support between the amounts called out in the last two rounds.
10 Having bids submitted electronically over the Internet, and using specialized software to administer the bidding process, has been used successfully in the FCCs spectrum auctions, as well
as many other successful auctions around the world.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00107

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

104
bidder from lying low in early rounds to conceal its intentions, or to allow rivals to eliminate one another.11 In areas where there are few bidders, the auctioneer may limit the information provided to each bidder. For example, each
bidder may know the number of other bidders, but not the identity of each.
The clock auction rounds continue until there is no more than one bidder for
each area.
At the end of the clock phase, there may be some areas for which there is no
bid. There may also be areas where bids have been submitted, but these do not
satisfy the aggregate reserve constraint because, as discussed above, the sum
of the wire center reserves will be greater than the aggregate reserve constraint
for the study area.
At the end of the clock phase, the auctioneer runs an optimization program that
selects the winning bidder in each area, based on all the bids submitted (this
may include bids from earlier rounds, since all bids remain in effect until the
auction closes). The optimization seeks to select winners for as many areas as
possible, while minimizing the cost to the fund.
d. Proxy Phase
Once the clock phase of the auction has been completed, a final round or proxy
phase is held to fine-tune the results.
The proxy phase is used to make the results of the clock phase more efficient. The
proxy format opens up additional bidding opportunities by allowing each bidder to
specify package prices that might not have been announced by the auctioneer in the
clock phase. At the same time, the proxy phase limits each bidders ability to behave
strategically by having a proxy agent bid on behalf of the actual bidder according
to strict rules.
In the proxy phase, each bidder reports a valuation for each package of areas in
which it is interested. This valuation is the best and final support amount that
bidder would accept. Unlike the clock phase, where each bidder specifies a single
package in each round, here a bidder may submit valuations for any number of
packages, and the packages may overlap in the sense that a given wire center may
be included in more than one package.
The actual bidding is then done on the bidders behalf by a proxy, which is simply
a computer program that bids according to preset rules, given the valuations submitted. Starting with the support amounts produced by the clock phase, each proxy
looks for opportunities to make its bidder better off by submitting a bid on the bidders best package; that is, the package that maximizes the difference between the
current bid and the bidders valuation. Bidding continues until no proxy can find
any such opportunity.
The same reserve rules discussed in Section 3.b are maintained in the proxy
round. The activity rule is also maintained in the proxy phase, but may be relaxed by a measured amount to allow bidders to increase the number of areas
on which they bid.
In practice, the proxy round is implemented using an optimization program. A
winner is chosen for each area by a criterion that minimizes the total amount
bid over all areas. The amount of support determined by the optimization is also
competitive in the sense that no coalition of bidders can offer the auctioneer a
lower-cost plan.
In the final optimization, there may still be some wire centers for which there
is no bid. There may also be study areas for which bids were submitted, but
where the auction fails because the bids did not meet the aggregate reserve constraint for the study area. In these areas, the situation would revert to the status quo prior to the auction, and the ETC(s) that participated in the auction
would continue to receive support capped by the mechanism described in Section 1.
The proxy phase builds upon the advantages of the clock phase. The information generated in the clock phase helps bidders formulate the valuations they
are asked to submit in the proxy phase.
If the areas are substitutes, the clock auction may also do most of the work of
identifying the best bids, leaving relatively little need for fine tuning in the
proxy phase.
11 Specifically, the rule employed here is called a revealed preference activity rule, which ensures that, as the support amount declines during the rounds of bidding, a bidder cannot shift
its bid toward a package whose support amount has fallen more than the support amount from
a previously preferred package. See Ausubel, Cramton, and Milgrom, op. cit., at page 120.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00108

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

105
However, where areas are complements, it is likely that bidders may hold back
from making some bids, and the clock phase may end before all of the possible
bids have been revealed. Suppose a bidder is interested in a package of areas
A, B, and C which it views as complements. Given the particular support
amounts called out by the auctioneer, and especially if another party bids aggressively for B, this bidder may choose not to bid for any of the three areas,
even though its combined bid might have been superior. By giving the bidder
an opportunity to specify a different combination of support amounts, the proxy
phase may elicit a bid for the package that would be better, from the auctioneers perspective, than any combination of bids offered in the clock phase.
7. Transition: Implementation of Auction Results
After the auction results have been announced, a transition period is necessary
if a winner will be taking on new Universal Service obligations. For example, if
the winner is a wireless CETC not already serving the area, then a transition period
may be needed. At some pre-announced point in the transition, the administrator
could require the winner to post bonds to ensure performance of the contract. Later
in the transition, the winner may be required to file an implementation plan to
show how it would plan to fulfill its responsibility. This would create an incentive
for the winner to formulate plans in a timely way, and would provide the administrator with an early warning of any potential problems. A transition period would
also allow ETCs that had participated in an auction, and had not won, to adjust
their business plans.
Transition in the Event of a General Auction. Under this proposal, no general auction would be held unless the Commission took action pursuant to its review in Step
4. If a general auction is held, and the ILEC is the winner, then no transition would
be needed, since the obligation it would take on would simply be an extension of
what it is already doing. If an ILEC bids for an area and loses, the state commission
would decide whether and how to reduce regulation of that carrier and what (if any)
obligation to serve would be appropriate. The Commission and/or state commissions,
on the other hand, could decide to exercise their authority to remove obligations
that the losing ILEC bidder may have to provide unbundled elements or resale.
Although the winner would have the responsibility to provide service, it could fulfill that responsibility by contracting with other parties, including the incumbent.
The losing ILEC could choose to continue to operate, selling retail services to endusers. The state commission may reduce retail regulation of such ILEC services. The
ILEC could also sell wholesale inputs to the new Universal Service provider. If the
FCC and/or the state commission removed UNE and resale obligations from the
ILEC, then these wholesale transactions could be at commercial terms.
8. Terms of the Contract
The contract between the winner and the regulators (FCC and state) would incorporate the terms of the RFQ and the level of annual support to the winner. Like
any procurement contract, it would include provisions to ensure that the terms of
the contract are met. These could include fines, forfeiture of bond amounts, and
being barred from participation in any subsequent auctions.
The contract would be awarded for a set term. The area could not be nominated
during that contract period. At the end of the term, the contract would continue
until a partyeither an ETC or the Commissionnominated it again, at which time
another auction would be held.
9. Areas Not Yet Auctioned
In some areas, support may not have been set through competitive bidding (either
because the area was not nominated for bid or because the auction failed to produce
a result). These areas would continue under the capped support arrangement described in Section 1. In an area that receives no support today, the reserve would
be zero, and thus that area would not be eligible for auction.

The CHAIRMAN. I thank you very much.


May I now call upon Senator Thune?
STATEMENT OF HON. JOHN THUNE,
U.S. SENATOR FROM SOUTH DAKOTA

Senator THUNE. Thank you, Mr. Chairman.


And I thank you for holding this hearing. And I apologizeIve
been bouncing back and forth between an Armed Services Com-

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00109

Fmt 6633

Sfmt 6601

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

106
mittee hearing this morningfor not being here for all the testimony. I thank our witnesses.
And I particularly want to welcome here a South Dakotan. Tom
Simmons is the Senior Vice President of Public Policy for
Midcontinent Communications, which is located in Sioux Falls,
South Dakota. And I happen to be a subscriber. They provide
phone, they provide Internet, they provide video. And so, I have all
those at my home in Sioux Falls. And I appreciate the efforts that
theyre making to improve those services all across South Dakota.
As you know, Im a cosponsor of Senator Stevens bill that deals
with USF, both on the distribution mechanism and the contribution mechanism, and think that thatcan improve the way that we
go about reaching some of these areas that arent reached and delivering broadband to more areas across the country. But I certainly welcome other thoughts about how to do that, whether or not
those are the best solutions or whetherthere are perhaps other
ideas that might be even better ones.
So, thank you Tom, for being here today, for making the trip out
from South Dakota, and to all our witnesses for their excellent testimony. And well look forward to moving forward with some legislation that will address this very important issue.
Thanks.
The CHAIRMAN. Mr. Simmons?
STATEMENT OF W. TOM SIMMONS, SENIOR VICE PRESIDENT,
PUBLIC POLICY, MIDCONTINENT COMMUNICATIONS

Mr. SIMMONS. Chairman Inouye, Vice Chairman Stevens, members of the Committee, thank you for inviting me to testify today.
My name is Tom Simmons. Im the Senior Vice President of Public Policy for Midcontinent Communications, a leading provider of
cable telecommunications services, including analog and digital
cable television, broadband Internet, long-distance and local telephone services. We serve approximately 200 communities in North
and South Dakota, western Minnesota, and northern Nebraska,
generally classified as small or rural. The size of our communities
range from densities of 5 to 116 homes per mile of cable plant, and
populations range from less than 30, in Barlow, North Dakota, to
our largest community, in Sioux Falls, home of Senator John
Thune, which has a population of more than 140,000.
Midcontinent launched its broadband Internet service over 10
years ago, on April 15, 1996, in Aberdeen, South Dakota. At that
time, we made a pledge to bring advanced broadband services to
as many customers as possible, regardless of the size of the community.
At the end of last year, we completed a project to rebuild our
cable plant to 750 megahertz or better in 50 more Midcontinent
communities, bringing our total number of upgraded systems to
over 156; and that serves 95 percent of Midcontinents customers.
Customers in these communities now enjoy over 150 channels of
analog and digital video programming, broadband, high-speed
Internet service, high-definition television, digital video recording,
and video on demand.
Id like to start by simply pointing out that the entry of cable operators into telephony is great news for consumers across America.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00110

Fmt 6633

Sfmt 6601

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

107
According to recent reports, cable phone customers are saving over
$10 a month on their phone bills, and the anticipated consumer
benefit from competition over the next 5 years will total more than
$100 billion. And cable operators like Midcontinent are increasingly
bringing benefits of their competitive phone services to rural areas.
As I testified last Congress before this Committee, Midcontinent
strongly supports the goals and purposes of Universal Service. We
believe that quality telecommunications services should be available to all regions of the country at just, affordable, and reasonable
rates. A strong Universal Service program is an essential component of national telecommunications policy, and we share the concern of policymakers, industry stakeholders, and the public, that,
in its current form, the Universal Service program is not sustainable.
But, while there is general consensus that all aspects of the system, including contributions, eligibility, and level of support, are in
need of reform, there are a wide range of views as to how the program should be restructured.
With respect to distribution-related Universal Service issues, we
believe it would be a mistake to make broadband services eligible
for USF distributions in areas that already have a broadband provider. Its unnecessary and profoundly unfair for the government to
subsidize a broadband competitor to Midcontinent or any other
broadband provider that has already stepped up to the plate and
answered the call to help close the digital divide. Subsidizing competition is also a waste of scarce resources that should be targeted
to areas where a market-based solution has not developed.
Also, the continued growth in the size of the Fund is a matter
of significant concern to the cable industry, for a very simple reason: these costs ultimately are borne by consumers.
Our industry supports efforts to reduce the burden of Federal
support programs by more efficiently distributing support. In particular, we believe that reverse auctions, if structured properly,
offer an opportunity not only to reduce the size of the Fund, but
also promote competition in high-cost areas by making support
available on a more equitable basis.
Turning to the current USF contribution mechanism, cable recognizes that reliance on the assessment of interstate telecommunications revenues virtually guarantees that the funding base will
continue to shrink. To address this, the cable industry has long advocated the adoption of a telephone-numbers-based contribution
mechanism, a simple, yet effective, reform that will sustain the
long-term health of this Fund while adapting to the evolving technology and economics of voice telephony.
As stated above, Midcontinent and the cable industry strongly
support the goals and purposes of the Universal Service program.
We recognize that changes are necessary to ensure its continued viability. We appreciate that the legislation introduced by Vice
Chairman Stevens would give the FCC the option of establishing
a numbers-based assessment scheme. And wed like to work with
this Committee to give priority to a numbers-based option and ensure that future assessments are not extended to broadband and
Internet services.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00111

Fmt 6633

Sfmt 6601

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

108
The imposition of new fees on broadband service at the same
time policymakers seek to encourage more widespread deployment
and service penetration would be counterproductive and would
raise the price of high-speed Internet services for current and potential broadband customers. It would also penalize those who have
worked diligently to deploy broadband to nearly every part of the
country.
Thank you, Mr. Chairman, for inviting me to testify today. Id be
happy to answer your questions, or those of the members of the
Committee.
[The prepared statement of Mr. Simmons follows:]
PREPARED STATEMENT OF W. TOM SIMMONS, SENIOR VICE PRESIDENT,
PUBLIC POLICY, MIDCONTINENT COMMUNICATIONS
Chairman Inouye, Vice Chairman Stevens and members of the Committee thank
you for inviting me to testify today. My name is Tom Simmons and I am the Senior
Vice President of Public Policy for Midcontinent Communications, a leading provider
of cable telecommunications services including analog and digital cable television,
broadband Internet and local and long distance telephone services. We serve over
200,000 customers in approximately 200 communities in North and South Dakota,
western Minnesota, and northern Nebraska generally classified as small or rural.
The size of our communities range from densities of 5 to 116 homes per mile of cable
plant and populations ranging from less than 30 in Barlow, North Dakota to our
largest community, Sioux Falls, South Dakota, which has a population of more than
140,000.
Midcontinent launched its broadband Internet service over 10 years ago, on April
15, 1996 in Aberdeen, South Dakota, and made a pledge then to bring advanced
broadband services to as many customers as possible regardless of the size of community. At the end of 2005, we completed a project to rebuild our cable plant to
750 MHz or better in 50 more Midcontinent communities bringing our total number
of upgraded systems to 156, serving over 95 percent of Midcontinents customers.
Customers in these communities now enjoy over 150 channels of analog and digital
video programming, broadband Internet service, high definition television, and digital video recording capability. Midcontinent Communications is also a certificated
local exchange telephone service provider in North Dakota, South Dakota, and Minnesota. Midcontinent first launched facility based circuit-switched telephony in
2000, and in the last year launched its first digital VoIP phone service in Mitchell,
South Dakota. Since then, weve rolled out digital phone services in a number of
additional communities throughout our service area and plan to continue the conversion of analog to digital telephony in many more. Midcontinent is a privately
held company that has invested, and continues to invest, substantial amounts of private risk capital to bring advanced services to our customers without the assistance
of public funds. Were proud of our ability to deliver the services our customers demand, which are no less than those demanded and expected in major metropolitan
areas.
As a provider of telephone service in rural America, Midcontinent strongly supports the goals and purposes of the Universal Service Fund (USF). We believe that
quality telecommunications services should be available to all regions of the country
at just, affordable and reasonable rates. In that regard, even prior to the Federal
Communications Commissions recent order requiring that all VoIP providers pay
into the USF, Midcontinent and all other cable operators offering voice telephone
serviceeither by way of traditional circuit-switched telephony or VoIPhave always contributed to the Universal Service Fund.
The entry of cable operators into the telephony marketplace is great news for consumers across America. According to a recent J.D. Power report, cable phone customers are saving over $10 a month on their phone bills. Based on the projected
growth of cable phone services, Microeconomic Consulting and Research Associates
recently projected that the total anticipated consumer benefit from competition over
the next 5 years will total more than $100 billion. And cable operators, such as
Midcontinent, are increasingly bringing the benefits of their competitive telephone
services to rural areas.
A strong Universal Service program is an essential component of national telecommunications policy and we share the concerns of policymakers, industry stakeholders and the public that, in its current form, the Universal Service program is

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00112

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

109
not sustainable. While there is general consensus that all aspects of the system, including contributions, eligibility and level of support are in need of reform, there are
a wide range of views as to how the program should be restructured.
With respect to distribution related Universal Service issues, we recognize the
value in preserving and promoting this program which provides funding to companies that serve areas where market forces historically might not have resulted in
all customers being served. These market forces, however, are not static. Improvements in technology, particularly the transition to IP-based equipment and services,
have made it possible for cable operators and other facilities-based competitors to
serve areas that previously might not have supported competitive entry. Similarly,
incumbent local exchange carriers increasingly are able to provide multiple services
(including DSL and video) over infrastructure previously used solely to provide telephone service. This transition to markets in which there is facilities-based competition for voice and non-voice services calls into question the need for continued government funding at historical levels, and may eventually permit the total elimination of high-cost support in at least some markets.
The continued growth in the size of the fund, however, is a matter of significant
concern to the cable industry for a simple reasonthese costs ultimately are borne
by consumers. Based on the anticipated growth of cable telephony services, and the
corresponding growth in the share of the program that will be funded by cable consumers, our industry supports efforts to reduce the burden of Federal support programs by more efficiently distributing support. In particular, we believe that reverse
auctions, if structured properly, offer an opportunity not only to reduce the size of
the fund, but also to promote competition in high-cost areas by making support
available on a more equitable basis. The challenge is to reduce the burden on consumers and promote competition, without sacrificing the level of service provided in
these areas today. We believe that an auction program can achieve these goals if
it incorporates the following requirements.
First, reverse auctions will only be effective and technology neutral if they cover
relatively small service areas (such as census block groups) rather than service
areas that conform to the boundaries of a particular type of service provider.
Second, minimum levels of service to be offered and obligations to be met by all
bidders must be established. This should include some sort of carrier-of-last-resort
obligation, which will ensure that the fundamental goal of providing service to all
consumers is met. Any facilities-based provider that commits to meeting these requirements should be eligible to participate in the auction.
Third, bidders should be required to offer services using their own wired or wireless connection to the end-user. Such a requirement will provide an important incentive for the construction of competitive networks.
Fourth, eligibility to participate in an auction and receive the resulting support
should be contingent on accommodating requests for interconnection. Incumbent
carriers should not be permitted to collect government funding for their networks,
while at the same time blocking competitive entry and foreclosing the introduction
of more efficient, innovative technologies that will provide the ultimate cure for
high-cost networks.
Fifth, there should be no guarantee of support such that an incumbent local exchange carrier or any other provider is made whole through a government subsidy
if they receive less support than they did before the introduction of auctions. Any
type of guaranteed support or other guaranteed revenue stream would completely
undercut the rationale for moving to an auctions-based system, which is to reduce
the overall amount of support provided by the program.
Lastly, for each area subject to auction there should be a fresh look on a periodic
basis. As technology develops and companies continue to expand their networks, the
amount of support needed to serve any particular geographic area should continue
to decline to reflect increased efficiencies.
We also believe it would be a mistake to make broadband services eligible for USF
distributions in areas that already have a broadband provider. Midcontinent shares
this Committees desire to ensure that all Americans, including those who live in
rural communities, have access to high-speed Internet service. As I stated at the
outset, Midcontinent has spent hundreds of millions of dollars to upgrade its facilities and deploy broadband services in rural communities. We did this without a government mandate and without a government subsidy. We did it because we want
to make certain that our customers have the same access to advanced digital technology as all Americans. We took the risk and invested private capital in order to
provide broadband services in the communities we serve. It is unnecessary and profoundly unfair for the government to subsidize a broadband competitor to
Midcontinent or any other broadband provider that has already stepped up to the
plate and answered the call to help close the digital divide.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00113

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

110
We recognize that some form of subsidy may be necessary to promote broadband
deployment in remote rural areas where no provider is currently offering a
broadband service and it is otherwise uneconomic to do so. The cable industry has
offered support for legislation that would offer loans or tax incentives to companies
that deploy broadband services in clearly defined and carefully targeted unserved
areas. But the government should take great care not to subsidize broadband in
communities where companies are already offering consumers broadband service.
Subsidizing competition is unfair and a waste of scarce resources that should be targeted to areas where a market based solution has not developed.
However, despite our support for government programs that target funding to
unserved areas, we would like to point out that any program that subsidizes private
entities to deploy broadband service is fraught with the potential for abuse. An example of such a program, though well intentioned, is the current Rural Utilities
Service broadband loan program. Loan money from this program is being used to
subsidize cable and phone competitors in markets where there are already two or
more broadband providers. This type of subsidized competition penalizes private entities serving those markets and discourages private investment in rural America.
In its September 30, 2005 report, the Office of Inspector General of the U.S. Department of Agriculture found that the RUS had not maintained its focus on rural communities without preexisting service, questioned whether the government should be
providing loans to competing rural providers when many small communities might
be hard pressed to support even a single company, and observed that the RUS, by
granting such loans, may be creating an uneven playing field for preexisting providers operating without government subsidies.
While government subsidies may be necessary to promote broadband deployment
in unserved areas, the cable industry does not believe that Universal Service Funds
are necessary to spur further broadband deployment. Broadband deployment in this
country continues to grow at a robust rate, with the number of consumers that have
signed up for high-speed Internet service in the U.S. far exceeding any other country in the world. The cable industry, for example, has invested over $110 billion
since 1996 in order to provide high-speed Internet access and other advanced services throughout the country.
As of June 30, 2006, the Federal Communications Commission reported that
based on company data, cable modem service was available to 93 percent of households that could access cable TV service and the phone companies Digital Subscriber Line (DSL) service was available to 79 percent of households who could access ILEC telephone service. Kagan Research reported even higher numbers, stating
that cable broadband service is available to more than 94 percent of all U.S. homes.
With private industry investing in broadband deployment like never before, and
the successful roll out of broadband and other advanced services across the country,
it does not make sense to undermine the Universal Service programs principle purpose of promoting the availability of affordable telecommunications services to all
regions of the country.
Turning to the current USF contribution mechanism, cable recognizes that reliance on the assessment of interstate telecommunications revenues virtually guarantees that the funding base will continue to shrink. An increasing number of companies offer consumers voice telephone service for a fixed monthly rate that does not
differentiate between local or long distance calls. Companies also offer bundled
packages of digital services that include voice telephony. Most consumer VoIP services are offered without regard to intrastate or interstate distinctions. The fact is
that interstate telecommunications revenues have been declining and are predicted
to continue declining for the foreseeable future. As the line between what is a local
and long distance call continues to blur, the existing USF contribution mechanism
will become increasingly obsolete which threatens the viability of the program itself.
The cable industry has long advocated the adoption of a telephone numbers-based
contribution mechanism, a simple yet effective reform that will sustain the longterm health of this Fund while adapting to the evolving technology and economics
of voice telephony. Using telephone numbers would be a relatively simple means of
determining who should contribute as well as when contributions were owed and in
what amount. There would be no need to apportion provider revenues into interstate
versus intrastate or to determine which portion of a bundled offering represents
interstate telecommunications. It would also make no difference whether a service
was defined as a telecommunications service or as an information service. Under a
telephone number-based system, all that matters is whether or not the service uses
a phone number. As such, a numbers-based system promotes competitive neutrality
among providers and technologies and ensures that no provider of a voice telephone
service is placed at a competitive disadvantage due to disparate treatment with respect to Universal Service Fund contributions.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00114

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

111
While a numbers-based approach would capture any service designed as a replacement for plain old telephone service (POTS), it would avoid assessments on a service
that might include a voice component. Few would argue, for example, that applications, or devices, where voice functionality is ancillary to the actual purpose of the
service or devicesuch as voice enabled gamingshould be assessed for USF purposes.
Some have expressed concern that a numbers-based system would collapse as proposals to map telephone numbers to Internet addresses, such as ENUM, become a
reality. However, ENUM requires that a subscriber have an active telephone line.
If someday in the distant future a non-number based system were developed and
widely implemented, the telephone number-based contribution mechanism could
easily be adapted, as some form of unique identifier or address will always be necessary to route various types of voice communications.
Mr. Chairman, the reality is that interstate telecommunications revenues are declining and will continue to decline. Conversely, an FCC staff analysis shows that
the number of active telephone numbers is expected to grow for the foreseeable future, from 554 million numbers in use in 2004 to nearly 600 million numbers in use
in 2007. Moving to a numbers-based USF contribution mechanism embraces this reality and will ensure the Universal Service Fund remains solvent well into the future. Furthermore, it would create a more predictable and equitable split between
assessments collected by providers of local and long distance telephone services, and
between residential and business subscribers. Residential telephone subscribers
would generally pay less under a numbers-based plan. Assuming an appropriate assessment amount, even most one-line households with low long distance usage
would pay less under a numbers-based system than they do under the existing
interstate revenue model.
As stated above, Midcontinent and the cable industry strongly support the goals
and purposes of the Universal Service program and recognize that changes are necessary to ensure its continued viability. We appreciate that the legislation introduced by Vice Chairman Stevens (S. 101) would give the FCC the option of establishing a numbers-based assessment scheme and we would like to work with this
Committee to give priority to the numbers-based option and ensure that future assessments are not extended to broadband and Internet services. The imposition of
new fees on broadband service at the same time policymakers seek to encourage
more widespread deployment and service penetration would be counter-productive
and would raise the price of high-speed Internet services for current and potential
broadband customers. It would also penalize those who have worked diligently to
deploy broadband to nearly the entire Nation.
Contrary to assertions that broadband is negatively impacting universal service,
the impact has been minimal at best. Most VoIP services, for example, already pay
into the Universal Service Fund and a number-based plan would, in any case, capture these services into the future. The assessment of broadband service is unnecessary to the goal of a stable, sufficient and predictable Fund. Instead, a numberbased contribution mechanism addresses the current problems with declining interstate revenues and bundling of services, and captures new technologies and protocols such as VoIP.
Mr. Chairman, Midcontinent supports the goal of the Federal Government to assure that all Americans have access to telephony and broadband services. We have
invested hundreds of millions to help that goal become a reality. We recognize that
government subsidies may be the only answer in some high-cost rural areas. However, any government program designed to promote broadband deployment must be
technology and provider neutral and carefully defined and targeted to only those
areas that lack broadband service. Furthermore, any such program must be subject
to the most stringent government oversight to ensure that government funds are allocated only to areas that are defined as unserved and are not used to subsidize
competition.
Thank you for inviting me to testify today. I would be happy to answer any questions you or the members of the Committee may have.

The CHAIRMAN. Thank you very much.


Id like to thank the panel for its patience waiting for us.
Id like to recognize Senator Thune for questions.
Senator THUNE. Thank you, Mr. Chairman.
I guess I would like to direct a question to Mr. Massey at Alltel.
In the NTCA testimony, it was pointed out that wireless needs
wires. And I guess Im interested in knowing whether you agree

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00115

Fmt 6633

Sfmt 6601

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

112
with that statement. And how should that statement impact our
thoughts about a reverse-auction proposal that combines wireless
and wireline into the same auction? Would combining them into a
single auction impact the incumbent wireline infrastructure thats
already being deployed?
Mr. MASSEY. Senator, thats a great question. And the construction of a network of all of our communications network is very complicated, and I will do my best not to use arcane terms.
All networks use what is called backhaul. Backhaul is the conduit through which traffic that originates in the last mile is channeled to then go through switches and so forth to reach the other
side, the other last mile. Its the garden hose in the middle of all
networks. Yes, wireless networksall wireless networks, to some
extent, use backhaul.
However, Ill say, we believe we pay, as we say in Arkansas, full
retail for access to those networks. It is one of our highest expenses, which is access to those wireline networks through which
traffic is routed. And Im not aware that the subsidy affects that
pricing by one penny.
Senator THUNE. Do wireless carriers need the same level of USF
support as wireline carriers?
Mr. MASSEY. Do we need the same level?
Senator THUNE. Yes. I mean, whats the
Mr. MASSEY. We believe
Senator THUNE. What is the
Mr. MASSEY.we do, yes.
Senator THUNE. What are the differences in how a wireless carrier spends those dollars, compared to a wireline carrier?
Mr. MASSEY. I can give you pretty good answers on the wireless
side. Part of the problem onand Im glad you raised the actualcost issue, because its one of those things that has been in a number of the filings. Just a couple of things on that. And the answer
is, we dont know. The fact is, is that we dontwe have a pretty
good handle on our costs. Were a public company. When we submit, for accountability for our USF funds, we give to the FCC, really for the world to know exactly where every dime we spend goes.
The problem is that on the wireline side, there is an intermediary
and a black box, Senator, and its very difficult for us to tell what
their actual costs are. So, its very possible, in my opinion, notwithstanding some of the thinking that may be out there, that our actual costs to extend service to a customer could be more than the
actual cost of a LEC, of a wireline company, to extend those costs.
We just dont know. And weve not, frankly, seen any factual basis
for any differentiation there.
The so-called identical support rule is one that compensates
competitive carriers for essentially the underlying wireline provider. And one point that Id like to make on that isand if you
see our written testimony, youll see it therethat weve provided
some factual basis for that. Supports not really identical. Its not
necessarily the same in a particular market. So, were not necessarily receiving the same dollars to build the network in a market that the wireline companies are receiving for that market.
The third thing is, is that we just ask that youd beware, as you
consider this so-called actual cost concept, the unintended con-

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00116

Fmt 6633

Sfmt 6601

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

113
sequences of paying one competitor a lower rate than paying another competitor. We think that its part and parcel of the so-called
competitive neutrality concept that really has caused the flourishing of both wireline and wireless networks in underserved areas.
So, thats what we ask for.
Senator THUNE. Mr. Crothers, lets say that a reverse-auction regime is put in place, for USF fundingand that Alltel wireless
wins with the lowest bid in your service areas. What do you think
happens, then, to your service offerings? I mean, do your folks pull
up stakes? Would they defer new infrastructure investments? What
happens in that type of a scenario?
Mr. CROTHERS. Senator, its almost impossible to tell. First of all,
as we go toward this reverse-auction concept I think that its going
to become rapidly apparent the fatal flaws that are involved with
it. Were talking about different technologies, were talking about
different service territories, were talking about different technologies having different capacities. Its almost impossible to ever
compare apples to apples and oranges to oranges. However, the one
thing that we can state today, state yesterday and tomorrow, is
that every one of a local exchange companys costs are approved.
They are, throughout South Dakota. They are, throughout the
world of independent telephone industry. If one receives Universal
Service funding, theyre done by form and approved by the Federal
Communications Commission. In the last 18 months, there have
been a tremendous amount of audits, and expanded auditing, of
local exchange companies.
And so, I think what youll see is, those are absolutely proven
correct for the wireline industry. So, if, in fact, they are correct,
and they are no longer available, a number of things are going to
happen. The number one is, prices would dramatically have to increase. That, of course, would force people off of the network. Number two is that companiesin many cases, the independent telephone companiesin South Dakota, virtually all locally owned
are no longer going to be able to invest in their networks.
So, to me, sirand I mentioned this at one point, and I believe
it was in the written testimonya reverse auction truly is a race
to the bottom. The less you invest in your network, the less you invest in your subscribers, the less that you will have, the more impacted that the people of America will be. And its going to be
disproportionally harsh on rural Americans.
Senator THUNE. Just one last question, if I might, for Mr. Simmons.
Tom, you had suggested that this only provides support for
broadband deployment in areas that currently dont have it. Do you
think there will be viable business models that will take advantage
of a program that youve described? And would Midcontinent participate in that sort of a program?
Mr. SIMMONS. In describing these programs are you talking
about price supports on the broadband side?
Senator THUNE. Right.
Mr. SIMMONS. Well, weve advocated that any level of support,
whether it be from the Universal Service Fund or even from the
Agriculture Departments Rural Utilities Service Funds, that they
be allocated to serving unserved areas.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00117

Fmt 6633

Sfmt 6601

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

114
We believe that might have been the original intent of the RUS
program, but we have seen something quite different happen across
our service areas, where, in fact, those funds were used to subsidize competition where service providers have been providing
service for some timenot only one, but, in some cases, two providerswhile still leaving a lot of areas unserved.
But, frankly, the areas unserved in our part of the country are
becoming either very remote or very limited. In the state of South
Dakota, for example, there are only two communities with a population of more than 200 people that do not have broadband service,
which I think is quite remarkable. We have a lot of competition in
the markets where we provide broadband service. In the 200 communities where we provide service, we have broadband competitors
in one-third of those, which is a pretty good number, since we serve
very small communities. And, again, much to the credit of the rural
telephone companies, they are, more often than not, our competitors in providing those broadband services.
So, again, I understand the need for Universal Service support
for those particular unserved communities, but maybe not in the
communities where theyre challenging us for subscribers. There
may be an area outside of that part of the community that requires
the help. Thats why I was intrigued by Mr. Landis statement, in
the earlier panel, about granular information, to understand what
this is really about, to clearly understand what is really going on
in those communities. And I have had the privilege of hearing Mr.
Landis testify at several NARUC meetings and certain seminars,
where he has greatly endorsed the marketing approach to providing services in communities. And, again, citing the local cable
companies and local telephone companies that provide those levels
of service.
Senator THUNE. My times up, I thank the panel for their testimony.
Thank you, Mr. Chairman.
The CHAIRMAN. Thank you very much.
Senator Klobuchar?
STATEMENT OF HON. AMY KLOBUCHAR,
U.S. SENATOR FROM MINNESOTA

Senator KLOBUCHAR. Thank you, Mr. Chairman.


And thank you, to the panel. I used to practice in this area, way
back, before I was a prosecutor. So, I was actually thinking about
all the times I had seen the USF language cited in briefs and
things like that. We look back, and looked at it, and it said, in the
Communications Act of 1934, that the fundamental purpose of the
Fund is to ensure that all the people of the United States have access to, a rapid, efficient, nationwide and worldwide wire and
radio communications service with adequate facilities at reasonable
charges. And so, thats why Im so concerned, representing a State
that has a large metro area, but also rural, about this digital divide. And, you know, Mr. Simmons, you were talking about the
communities in South Dakota that have access, but the issue has
been acknowledged, is that we dont really have a tracking to know
how many people have it. The FCC tracks it by Zip Code, so one
person could have it within the zip code. But there was a study

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00118

Fmt 6633

Sfmt 6601

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

115
the GAO reported, in May of 2006, that broadband take-up rates
were 70 percent higher in suburban and urban homes than in rural
homes. And a 2006 Pew study found a similar divide. And one
more troubling statistic, more than one in ten rural counties dont
even have a single high-speed Internet connection in the entire
county.
So, with that, I just wanted to explore a little more about how
we can get to where we want to go. Mr. Massey, you have stated
in your testimony that wirelessI heard you say thisis often the
only means of high-speed broadband access in rural areas. Could
you talk about the implications for our national policy if we were
to go that way?
Mr. MASSEY. Well, in fact, Senator, we bought a company, Midwest Wireless, thats now a part of Alltel that was headquartered
in Mankato, Minnesota, the southern, more rural part, as you
know, and they have a vibrant wireless broadband practice. It delivers speeds that are close to DSL speeds. They werethey are,
and were, selling it very well. But, frankly, there are a number of
markets in thereally, in the more rural parts of southern Minnesotaas you know, a lot of farms, a lot of distanceand not a
lot ofand maybe, I guess youd almost saytheres a lot of lowincome population. Its not profitable for us. We dont get a return
for our shareholders to build a fully deployed broadband network
in some of those rural southern Minnesota markets. Wed love to
do it. Wed love to serve those customers. With a little help, we
could do that. And we think that we could do that as efficiently as
anybody at the table.
So, we think wireless is not theis not the sole answer to the
broadband problem. I think it wasVice Chairman Stevens was
talking about the call-centeropportunity to return a lot of callcenter jobs to rural America, lets say. Frankly, my guess is that
the technology to really move that sort of traffic will alwayswell,
at least for the foreseeable future, bewill be some sort of a
ground-based technology. Itll be fiber of some kind. It would be
very difficult and very expensive for us to build aand probably
the technology doesnt exist to build that fat a pipe for that sort
of service. But for the people in your markets and the people in
South Dakota and Arkansas and Tennessee that are agriculturebased, mining-based, farming-based, that want access to rapid
dataand where they are in their jobs, not to have to get in the
pickup and drive all the way back to the house to access the Internet, to order a combine part, but to get it where they arewe think
wireless is the solution, and we think it should be a part of any
broadband solution.
Senator KLOBUCHAR. Thank you.
Mr. Tauke, I understand that many rural wireless carriers that
Mr. Massey was referring to would like to provide broadband, but
they cant do it effectively unless they get data-roaming agreements
with larger carriers. I understand that you have a pretty standard
voice roaming agreement with the smaller carriers. Does your company make it common practice to enter into data-roaming agreements? And what would you say about a policy of automatic roaming agreements?

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00119

Fmt 6633

Sfmt 6601

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

116
Mr. TAUKE. Roaming agreements, for both voice and data, have
generally been commercial agreements. We try to enter into agreements, where we can, that make sense in order to have as expanded a coverage as feasible for our own customers, because, obviously, when we enter into a roaming agreement with someone else,
that gives us the ability to provide additional service to other customers.
There are not many wireless carriers inmany of the smaller
wireless carriers todaywho deploy the kind of data capability that
we do in our network.
Senator KLOBUCHAR. I mean, you also talked about how
broadband should not be a part of Universal Service, in your view,
because the issue is not affordable rates, I think you said, but, instead, the one-time capital investment. And I just wondered if some
of the other panelists could comment on that, and where you stand
on that.
Mr. Crothers?
Mr. CROTHERS. Senator, we believe that broadband should be
part of Universal Service. Its been demonstrated over and over
the first panel emphasized itthat it really goes to the security
and the competitiveness and the education of the American people.
It isnt a luxury, it isnt an add-on. And I know the time is getting
late, but Ill leave it at that. Its critical to our very being.
Dr. STAIHR. First, if were going to support broadband, its a
given that the Fund would actually have to increase. OK? If we
make that decision, we make that decision, and thats a good thing.
It comes down to, do the benefits of supporting it outweigh the
costs? And we know there are areas that, just as theyre uneconomic to serve for voice, they are going to be uneconomic to serve
for broadband without some help, regardless of the technology.
Maybe wireline, maybe wireless. As an economist, I think the data
is pretty clear that the benefits do outweigh the costs. So, a policy
that supports broadbands inclusion as a supported service makes
sense.
Mr. MASSEY. I think you got my answer earlier, but
Senator KLOBUCHAR. Yes.
Mr. MASSEY.just to make sure, we believe it ought to be part
of that Universal Service Fund.
Senator KLOBUCHAR. OK.
Mr. Simmons?
Mr. SIMMONS. Senator, my comment would be, again, that granular component, and take a look at what it really is. Mr. Crothers
said its not entertainment. Yes, it is. IP video is a major portion
of that. Lots of companies will make lots of money by providing almost a cable service over the broadband side of all that. The gaming components, and the time that are spent with those type of
things that fall clearly under the entertainment side, might be
something quite different than pure information flow or pure communications or Voice over Internet telephony.
Senator KLOBUCHAR. So, you would support it for certain components of it.
Mr. SIMMONS. Well, I think it would be important to take a look
at what is being subsidized and clearly understand what that service is used for, and if it really does merit support. Or if we have

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00120

Fmt 6633

Sfmt 6601

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

117
a case of unintended consequences where, down the road were subsidizing a service that we shouldnt, which puts at risk someone
elses private investment.
Senator KLOBUCHAR. And, you know, my concern is just that
were shipping jobs to other countries that have broadband available, and then we have small towns in Minnesota, where they dont
have it available, where we could add to their employment if we
did.
Mr. SIMMONS. Im not
Senator KLOBUCHAR. And I dont think of that as videos and entertainment.
Mr. SIMMONS. Im not questioning the need in those areas that
are unserved. I clearly think we need to do what we need to do to
make sure that broadband is deployed into those particular areas.
Senator KLOBUCHAR. Mr. Tauke? Thats my last question, then.
Go ahead.
Mr. TAUKE. Senator, I just wanted to clarify our position. We believe that we need to ensure that broadband is available to all consumers. Second, we recognize that there are areas where there is
no broadband today, and where there is need for assistance in
order to provide that broadband. But the question is, Do you want
a program which provides ongoing sustained funding, which is
what you have with Universal Service, or do you need a program
which provides significant capitalone-time capital investment?
We think the latter is whats needed now. And so, you cant look
to the Universal Service Fund as a solution to that particular problem.
Senator KLOBUCHAR. All right. Thank you for clarifying that.
[The prepared statement of Senator Klobuchar follows:]
PREPARED STATEMENT

OF

HON. AMY KLOBUCHAR, U.S. SENATOR

FROM

MINNESOTA

Thank you Mr. Chairman and Mr. Vice Chairman. I am pleased to be here to address the challenges facing the Universal Service Fund and to work on ways to reform it.
The USF is not new to mein my years as a telecoms lawyer I dealt frequently
with USF issues. And in the past two years, Ive been all over my home state, Minnesota, talking about the need to serve all of our communities with affordable and
up-to-date telecommunications services.
But I am a newcomer to the more recent debates about the best methods to sustain, reform, and fairly allocate the costs of the Fund.
So I look forward to engaging with all of the stakeholders, with the FCC and the
Joint Board, and with my colleagues, and to asking a number of important questions. They include:
Should the contribution be assessed on a per connection basis or a per working telephone number basis?
How do ETCs fit into the purposes and operation of the Fund?
How do we best allocate high cost support for non-rural carriers?
How do we improve our method of distribution from the Fund?
Throughout this debate, I think it is vital that we remember the fundamental
purpose of Universal Service, as stated in the Communications Act of 1934: It is to
ensure that all the people of the United States, have access to a rapid, efficient,
Nation-wide, and worldwide wire and radio communication service with adequate
facilities at reasonable charges . . .
And that brings me to my top priority in this area: bridging the digital divide and
bringing high-speed broadband to every community in Minnesota and every corner
of this country.
I have talked in previous hearings about the persistent urban-rural digital divide.
In May 2006, the GAO reported that broadband takeup rates were 70 percent high-

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00121

Fmt 6633

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

118
er in suburban and urban homes than in rural homes. A 2006 Pew study found a
similar divide.
Here is another troubling statistic: more than 1 in 10 of the most rural counties
do not even have a single high-speed Internet connectionin the entire county.
A community that is left without affordable broadband access is a community that
will be left behind. A 2006 MIT study found that towns which had mass-market
broadband experienced markedly faster growth in employment and number of businesses.
I am convinced that the market alone will not solve this problem. Broadband deployment will lag behind in rural areas because the private sector gets a much higher return in areas of high population density and high income. I am convinced that
the Federal Government must assist underserved areasespecially rural areasin
partnership with states, towns, and the private sector.
That much is clear to me. What is a little less clear is the precise form Federal
Government that involvement should take.
The FCC and the Joint Board have resisted adding broadband to the list of covered services under the USF. They have consistently decided not to. I want to explore that decision, especially in light of the fact that broadband meets one of the
key criteria of the Fund: like plain-old telephone service decades ago, it has become
essential to education, public health, and public safety.
I believe that an updated, reformed Universal Service Fund is very likely the best
vehicle for bringing broadband to rural America. But I am willing to listen to those
who say that some other vehicle will get us faster and more effectively to our destination.
Some have talked about the possibilities of auctions, others about targeted grants
and loans, others about tax credits. And, of course, any Federal approach must complement existing and emerging digital divide initiatives being undertaken at the
state and local level.
I intend to look at every possible vehicle, with a strong inclination toward adding
broadband to the USF, and I do not intend to rest until we have met this challenge.
Thank you.

The CHAIRMAN. I thank you very much, Senator.


Id like to stay here a little longer to continue this discussion, but
Im already an hour late. Im supposed to be on the floor right now.
But Ive just instructed the staff to add, to the in-depth briefing,
wireless.
Im certain you have noted that the membership of this Committee is heavily rural. And, as a result, I can assure you that we
will have some action here, if not the consideration of some measure on the floor. But in order to do that, we will have to take into
consideration the concerns and interests of all the parties involved.
It will be a challenge, but I can assure you we will take on that
challenge.
And, in the meantime, well be calling upon you for advice and
counsel, because this is not the way to do policy and make decisions. A whole bunch of nonexperts here, we know very little about
what is involved, but we will have to make the decisions. And so,
we are counting on you.
And, for the moment, Ill be submitting questions for your consideration. Id like to get a better understanding of reverse auction,
for example. Id like to get your thoughts on that.
So, with that, thank you very, very much.
The hearing is adjourned.
[Whereupon, at 12:20 p.m., the hearing was adjourned.]

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00122

Fmt 6633

Sfmt 6601

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

A P P E N D I X
PREPARED STATEMENT

OF

HON. BILL NELSON, U.S. SENATOR

FROM

FLORIDA

Mr. Chairman, thank you for holding this important hearing on the present and
future of the Universal Service Fund.
Today, the Universal Service Fund fills a critical role by ensuring that all Americans have access to telecommunications services at affordable prices. The E-Rate
program, for example, has ensured that almost all American students have access
to the Internet. Similarly, the Low-Income program provides affordable telephone
service that is truly a Lifeline for many families.
As we move toward the future, I look forward to exploring possible new uses of
Universal Service funds, such as targeted support to bridge the urban-rural divide
in broadband service penetration. Consumers in rural areas of Florida should have
the same access to broadband services that consumers in urban areas, such as
Miami or Tampa, have available.
At the same time, however, we must also take steps to preserve the financial stability of the Universal Service Fund. This reform should start with controlling the
growth of the Universal Service Fund and, in particular, growth of the High-Cost
portion of the Fund.
Unrestrained growth of the High-Cost portion of the Fund is causing an increased
and substantial burden on consumers that pay into the Fund. Florida, for instance,
is currently the largest net payer into the Fundlast year the State paid in more
than $311 million more in contributions than it received in distributions.
We need to move toward a system that shares both the costs and benefits of Universal Service more equally among all Americans.
On the contribution side of the Universal Service equation, I look forward to reform that is sustainable, while still protecting low-volume and low-income consumers from any spike in the amount they currently pay into the Fund.
And on the distribution side, I look forward to hearing the details of various reform optionssuch as reverse auctionsthat may limit unsustainable growth of the
Fund.
Working together, we can create a Universal Service Fund that is technologically
flexible, fair to consumers, and sustainable for the future.
Thank you.

PREPARED STATEMENT OF GARY WALLACE, VICE PRESIDENT,


CORPORATE RELATIONS, ATX GROUP, INC.
Mr. Chairman, Vice Chairman and members of the Committee, on behalf of the
ATX Group, Inc., thank you for giving ATX Group the opportunity to submit comments for the record to addresses the devastating effects that a one size fits all
numbers-based Universal Service Fund (USF) contribution structure would have on
consumers with automobiles equipped with integrated in-vehicle emergency communications systems.
I applaud the Senate Commerce Committee for specifically recognizing in its 2006
proposed telecommunications law reforms that the unique circumstances of in-vehicle emergency communications should be taken into account in designing a Universal Service Fund contribution mechanism.
In-vehicle emergency communications systems which provide automated crash notification, stolen vehicle recovery and mayday signals to trained emergency response
professionals are often referred to as telematics services. These intelligent vehicle
technologies enhance response to highway emergencies. Every day these services
save lives, speed emergency response and assist drivers. Systems deployed today on
several million passenger vehicles provided by ATX, OnStar and others use the cellular network with communications devices which have individual telephone numbers. These systems, however, are extremely low volume network users.
(119)

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00123

Fmt 6601

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

120
ATX currently pays Universal Service support payments through its carrier suppliers and for telecommunications services provided by its call center. As a matter
of principal, ATX has no objections to making equitable contributions to the Universal Service support system. The proposal to assess a flat Universal Service fee
on all telephone numbers, regardless of level of use would have a profound, inequitable and burdensome affect on lifesaving telematics services.
Even at $1.00 per month per phone number, the USF contributions for telematics
services would approach the cost of the telecommunications services charged by the
carrier. The fee would violate the Telecommunications Act of 1996 requirement that
Universal Service contributions be equitable and nondiscriminatory. It would also
be against the public interest to slow the broader rollout of telematics based safety
and security services to the mass auto market. This would be contrary to the long
held transportation and public safety policies of encouraging drivers to adopt intelligent transportation technologies.
ATX provides core automotive telematics services to several auto Original Equipment Manufacturers (OEMs). Core telematics services include GPS satellite location-enhanced, automatic collision notification; a dedicated in-vehicle May Day
button to summon emergency assistance; and vehicle theft recovery. Neither ATX
nor its automotive OEM customers currently offer a personal calling service as part
of their telematics packages. The core service allows a vehicle occupant to communicate with a call center to request assistance.
Additionally, upon deployment of a vehicles airbag and/or activation of emergency
pretensioners in seat belts, a signal is transmitted to the call center, which will respond to the automatic crash notification (ACN). Whether by call or ACN signal, a
vehicles transmission is only to the ATX call center and only the call center may
place a call to the vehicle. The technology uses the cellular network, with GPS location capability, and each activated vehicle is assigned one telephone number.
An overwhelming number of vehicles have no communication with the call center
during a year. Of those who do communicate with the call center the average call
is of very short duration. The presence of a phone number reflects neither network
use nor the ability to communicate outside the call center and vehicle. A consumer
purchases core telematics services to summon assistance in an emergency. A one
size fits all numbers-based USF assessment will have substantial impact on consumer behavior, will encourage arbitrage opportunities between mobile communications technologies and be damaging to the effort to bring emergency communications
capabilities to all vehicles.
In a hypothetical 350,000 telematics equipped vehicle fleet, a $1.00 per month fee
against each telephone number results in a USF contribution of $350,000 per
month. The current USF fee for the same fleet would be approximately $10,000. The
proposed USF assessment approaches the cost paid for the airtime and the underlying services provided by the carrier. Notably, the carriers services encompass not
only airtime, but its expertise and administrative assistance in assigning numbers,
arranging for toll free platforms, initializing a vehicles capability to transmit and
receive, maintaining databases and overall assisting in the delivery of emergency
telematics services.
The Communications Act of 1934, as amended, section 254(b)(4), establishes the
standard by which the FCC then may assess a fee to support the Universal Service
program. That standard requires that the contribution be equitable and nondiscriminatory. A contribution mechanism that approaches the cost charged and
revenue collected by the carrier for its services clearly violates this standard.
The courts have addressed the importance of how the fee must be fair. In Texas
Office of Public Utility Counsel v. FCC, 183 F.3d 393, 431 (CA 5 1999), the Court
of Appeals for the Fifth Circuit ruled that a Universal Service fee that exceeded a
carriers revenue violates the laws equitable and nondiscriminatory standard. The
Court held that where a carrier was assessed a fee in excess of its interstate revenues, the underlying premise required of any contribution mechanism was violated.
There must be fairness in the allocation of contribution duties. It characterized the
assessment as a heavy inequity and that the cost imposed was prohibitive.
The Fifth Circuit addressed the circumstances where a carrier had minimal interstate traffic and significant international traffic. The core telematics circumstance
is even more egregious. Here, with the ability only to communicate between call center and vehicle, and where most consumers make no calls, network use is nominal
and confined. The fundamental value of telematics is the ability to transmit a call
or signal to the call center in those infrequent circumstances when emergency assistance is needed. The current USF contribution model, based on revenues, recognizes and accommodates the vast disparity between general consumer use of the cellular network and the minimal use of core telematics equipped vehicles.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00124

Fmt 6601

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

121
A one size fits all phone number assessment structure does not comprehend that
while automotive telematics services are assigned a large number of phone numbers, the extent and frequency of use of the network is extremely low and confined.
The FCCs own decisions recognize that a contribution model must recognize and
accommodate such disparity. See In the Matters of the Federal-State Joint Board on
Universal Service and Access Charge Reform, 15 FCC Rcd 1679, FCC 99290 at
paragraphs 2325 (1999).
A $1.00 monthly fee on each telematics vehicle is inequitable and discriminatory.
Under this assessment, ATXs customers would see their monthly USF contributions
increase nearly 3,000 percent, approaching the cost of the wireless service. Even
under the 50 percent discount proposed by the cellular carriers for their buckets
of minutes customers, where several numbers are assigned yet only one bill is rendered, the proposed USF fee to core telematics vehicles is still enormous. Such an
assessment will disrupt a market that today is delivering an important public safety
featurethe ability to locate expeditiously and dispatch aid to individuals involved
in an in-vehicle emergency or collisionubiquitously and without limitation to the
technical capabilities of local Public Safety Answering Points.
If the expansion of location-based automatic crash notification and emergency response services are slowed, it will profoundly affect rural areas where these services
have the greatest impact on highway deaths and injuries. Because distances are so
great, the speed of emergency response in a rural setting is the difference between
life and death as well as recovery and permanent injury.
In summary, ATX urges the Congress and the Federal Communications Commission to recognize what the Senate Commerce Committee recognized last year. A one
size fits all numbers based systems is profoundly unfair and inequitable to drivers
of vehicles equipped with integrated in-vehicle emergency communications systems.
PREPARED STATEMENT OF F.J. POLLAK, PRESIDENT
TRACFONE WIRELESS, INC.

AND

CEO,

My name is F.J. Pollak. I am President and Chief Executive Officer of TracFone


Wireless, Inc. TracFone is headquartered in Miami, Florida. With more than 8 million customers, TracFone is the Nations leading provider of prepaid wireless telecommunications services and TracFone is also the 6th largest wireless carrier in the
United States. (The only larger wireless carriers are AT&T/Cingular, Verizon Wireless, Sprint Nextel, T-Mobile, and Alltel). Since its inception in 1996, TracFone has
been able to grow its business to over 8 million customers by focusing on a segment
of the wireless marketplace largely ignored by other wireless companies. Specifically, TracFones service is directed mainly to low volume, often low income, consumers who normally make an average of 1 call a day. TracFone offers a pay-asyou-go service. There are no duration or volume commitments, no early termination
penalties, no advance deposits; no credit checks. TracFones customers pay only for
the wireless service they need, when they need it. For many TracFone customers,
wireless telephone service would otherwise be unavailable or, if available, would be
unaffordable. As such, TracFone thinks of itself as a true Universal Service Providerand it provides affordable, easy-to-use prepaid wireless service without receipt of any subsidies from the Universal Service Fund.
As a provider of interstate telecommunications services, TracFone is required to
contribute to the Federal Universal Service Fund (USF). Although TracFone contributes to the USF based on its actual interstate revenues, it has no way to recover
its USF contribution costs from consumers in the form of billed surcharges. Unlike
traditional providers of post-paid wireline and wireless services, prepaid providers
do not send monthly invoices to their customers and therefore, have no opportunity
to add Federal Universal Service Fund surcharges as line items on customer bills.
With no means to recover its USF contributions from its customers, today, TracFone
contributes over $10 million a year into USF out of its shareholders pockets. As
such, TracFone is a substantial contributor to the Fund and is a very meaningful
voice in the USF debate.
TracFone believes that the current USF contribution methodology based on interstate revenues is fair to all and is consistent with the legal requirements of the
Communications Act. To the extent that there are concerns about the ability of the
current, interstate revenues-based system to provide sufficient support for the USF,
TracFone believes that certain adjustments could significantly increase the level of
USF funding. Specifically, there no longer is any need for a wireless safe harbor as
wireless providers are able to identify which of their usage is interstate. In that regard, TracFone believes that the FCC took an important step in the right direction
last June when it increased the wireless safe harbor from 28.5 percent to 37.1 per-

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00125

Fmt 6601

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

122
cent. In addition, TracFone supports the decision of the FCC to subject Internetbased telephone calling services (often called Voice over the Internet Protocol or
VoIP) to USF contribution requirements. Also, the law empowers the FCC to impose USF contribution obligations on others who provide services which use interstate telecommunications including, for example, broadband Internet access services. TracFone believes that the contribution base could be expanded to include
those services with no reduction in demand for those services. Inclusion of providers
of broadband Internet access services in the USF funding mechanism seems especially appropriate in light of proposals which would expand the USF to subsidize
such services in high-cost areas.
A contribution methodology based on working telephone numbers would significantly and unnecessarily increase the costs of service for low volume low income
consumers. Today, based on its actual interstate revenues, TracFone remits to the
USF about $0.10 per customer per month. While this may seem like a small
amount, TracFones average revenue per user is only $14.00 per month as compared
with the wireless industry average of about $56.00. Moreover, TracFone customers,
like most prepaid wireless customers, make few interstate calls. Indeed, many
TracFone customers make no interstate calls. Therefore, almost all of its customers
$14.00 average revenue is derived from intrastate and local serviceservices which,
by law, may not be subject to assessment for the Federal USF.
If the FCC were to implement a numbers-based contribution methodology and the
initial per number charge were to be set at $1.20 per month (an amount projected
by a group called the USF By The Numbers Coalition in a January report),
TracFones monthly per customer USF contribution would increase from $0.10 to
$1.20more than a 1,200 percent increase, effectively creating almost a $100 million a year tax increase. Since TracFones customer base has grown rapidlyby approximately 1.8 million customers in 2006, future increases under a numbers-based
plan would be much greater. As discussed above, TracFone has no means to recover
USF contributions from its customers through billed surcharges, TracFone would
have to absorb the entirety of these increases from its operating revenues since raising its rates is not a viable, competitive option.
The reason why TracFone and other prepaid providers cannot raise their rates to
incorporate their USF contribution costs is the nature of the competitive market in
which telecom services in general, and wireless services in particular, are provided.
Traditional post-paid providers (those who render bills for services) widely advertise
the price of their services without reference to USF surcharges or other additions
to those advertised prices. Such carriers widely advertise services such as 400 minutes for $39.95, or $0.10 per minute, etc. However, their bills sent in arrears are
for much higher amountsamounts which include USF and other taxes, surcharges
and fees imposed by the carriers but not included in their advertised price. Companies like TracFone compete with those providers. Unlike post-paid providers who
can add taxes, fees and surcharges, including USF charges, to their advertised
prices, TracFone and other prepaid providers must include in their advertised prices
all taxes, fees, and surcharges, since they have no billing mechanism to add those
charges later. This creates a significant competitive disadvantage since consumers
compare providers advertised prices with each other, without realizing that some
providers advertised prices do not include taxes, fees and surcharges which will be
added to their bills, while prepaid providers advertised prices are all-inclusive.
Some providers of prepaid servicethose who provide service using their own
switchesare able to take from their customers prepaid account balances usage
amounts equivalent to the amount of the USF surcharge. This method is often
called the Sufficient Positive Balance method since the providers will debit the
customers accounts only if there are in the accounts a sufficient positive balance
to cover the amount of the debit. Unlike those providers, TracFone does not have
any switches of its own. It provides service by purchasing capacity from other providers. As a result, TracFone customers account balances are stored directly in the
customers wireless phones, not in a central switch. TracFone does not have real
time access to its customers phones or to the prepaid account balances stored in
those phones, and it could not debit those accounts to recover its USF costs even
if it wanted to. Accordingly, neither raising rates nor debiting customer accounts to
recover USF costs are viable options for providers like TracFone. In short, a numbers-based contribution plan would not work for certain types of telecom providers,
including prepaid wireless providers. Not only are those companies services not
billed, those companies do not provide service on a monthly basis. Some consumers
make multiple purchases of prepaid airtime in a month; other consumers may go
several months or more without making any airtime purchases.
Consumer groups have recognized that a numbers-based plan would dramatically
increase the costs of the USF borne by low income low volume consumers. That is

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00126

Fmt 6601

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

123
why a coalition of such groups called the Keep USF Fair Coalition as well as the
National Association of State Utility Consumer Advocates has opposed the implementation of a numbers-based contribution proposal at the FCC. In a report released February 27, 2006 entitled Exposing the Hoax: The Phony Crisis of the
Universal Service Fund, the Keep USF Fair Coalition articulated the view that a
flat per-working telephone number tax would significantly increase the monthly
telecommunications costs for low volume consumers and would force many low income consumers to drop their telephone service. The Coalition report also demonstrated that abandonment of a revenues-based system in favor of a numbers tax
is not necessary, pointing out that the contribution base has been stable and that
available data demonstrate that there will not be a sharp decline in interstate telecommunications revenues.
Moreover, the potentially devastating impact of a regressive numbers tax to finance Universal Service is not limited to residential consumers. Many so-called enterprise customersusers of large quantities of telephone numberswould also be
hit hard by a numbers tax. One prominent example of such users is the higher education community. The FCC has heard from numerous colleges and universities,
large and small, about how their telecom costs will increase dramatically if a per
number tax is implemented. For example, Harvard University estimates that its annual USF contributions would increase from $70,000 to $400,000; Rice University
anticipates monthly increases from $400 to $10,000; Southern Illinois estimates that
its annual USF fees would increase from $12,000 to more than $200,000 per year;
Calvin College, a small liberal arts college in Michigan, would have its monthly USF
costs skyrocket from $700 to over $11,000. The list goes on.
These institutions differ from each other in many respects. However, the ability
of each institution to provide telecommunications services to its students and faculty
would be undermined by the FCC numbers tax proposal. Several (including Harvard) even report that their ability to provide E911 access for their students would
be jeopardized. Given the high priority which the FCC properly has placed on mandatory E911 access availability, it would be a sad and cruel irony if the FCCs
numbers tax had the perverse impact of limiting E911 access for students residing
on college campuses throughout the country.
There is another problem with a numbers tax. Typically, telephone numbers are
provided as part of local telecommunications service. Many customers of wireline
and wireless telephone service make few, if any, interstate calls. Yet the FCCs proposed monthly numbers tax to finance the Federal Universal Service Fund would
be imposed on such customers without regard to whether consumers derived any
interstate usage whatsoever in any given month. Imposition of USF funding obligations on such consumers was not what Congress had in mind in enacting Section
254 of the Communications Act; nor would it be sound public policy to require that
consumers who use little, if any, interstate service, bear a largeand increasing
share of underwriting the Federal USF.
If the FCC adopts a numbers tax to fund universal service, it will be necessary
for it to provide alternative contribution mechanisms for certain types of carriers.
Many providers of interstate telecommunications service do not provide customers
with working telephone numbers as part of their service offerings. Since the law requires that every provider of interstate telecommunications service must contribute to the USF, there must be a mechanism appropriate for all carriers.
TracFone recommends that those interstate telecommunications service providers
who are unable to recover their USF contributions through billed charges to their
customers be allowed to continue to have their contributions based on their interstate revenues. Alternatively, in order to prevent pricing their services beyond the
reach of the low volume, low income users they serve, TracFone suggests that those
carriers USF contributions under any methodology be capped at the levels of their
contributions under the current revenues-based methodology.
Finally, TracFone reminds the Committee that another component of the efforts
to ensure that USF contributions not unduly burden the provision of telecommunications services is to demand that the Funds growth be limited and distribution of
USF resources carefully managed. TracFone urges the Committee to continue to encourage the FCC to protect against waste, fraud and abuse, and other sources unintended and avoidable growth of the USF. In this regard, TracFone believes that the
most critical Universal Service issue is the rising size of the Fund and the increasing burden being borne by the Nations telecommunications consumers to support
that unrestrained growth. The use of reverse auctions as a means for distributing
USF high-cost support has the potential to significantly limit growth of the fund.
The reverse auctions proposal is currently before the Federal-State Joint Board on
Universal Service and a recommendation to the FCC is expected some time this
spring. TracFone encourages the Joint Board and the FCC to give careful consider-

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00127

Fmt 6601

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

124
ation to reverse auctions and other proposals to limit fund growth and to ensure
that USF resources are distributed in an efficient manner. Congress and the FCC
must enact and implement requirements and procedures which limit availability of
USF support to those who truly need the support and which ensure that the funds
are disbursed in an efficient and targeted manner, with safeguards to prevent
waste, fraud and abuse. Implementation of such requirements and procedures will
ensure that there will be a sufficient USF in the future without the need for disruptive and inequitable numbers taxes imposed on consumers and on those enterprise
customers, including colleges and universities and healthcare institutions, which
utilize large quantities of phone numbers.
GOLDEN WEST TELECOMMUNICATIONS
March 8, 2007
Hon. DANIEL K. INOUYE,
Chairman,
Senate Commerce, Science, and Transportation Committee
Washington, DC.
Dear Senator Inouye:
On behalf of Golden West Telecommunications, I commend your leadership in the
Senate Commerce, Science, and Transportation Committee. I have watched with admiration as you and other members of the Committee work to ensure the long-term
stability of the Universal Service Fund. While policy positions of small and large
companies may differ on this issue, we are in agreement that accurate, complete
and factual testimony is the foundation for sound public policy. Unfortunately, we
do not believe this goal has been met.
During his testimony on March 1, 2007, Richard Massey, who currently serves as
Executive Vice President, Corporate Secretary and General Counsel of Alltel Wireless, testified about the successes that Alltel has had across rural America. In his
pre-filed testimony Massey stated:
For example, on the Pine Ridge Reservation in South Dakota, the tribe estimated that less than 30 percent of the population had telephone service prior
to Alltels entry into the market as a wireless Universal Service provider. Today
more than 80 percent of the population on the Pine Ridge Reservation has access to wireless telephone service.
In addition, during his oral statement before the Committee, Massey stated:
Its Pine Ridge Reservation, its included in one of the poorest counties in the
United States. When we found this market some years ago, it included an incumbent wireline provider that receives Universal Service funds, and yet only
20 percent of the population on this reservation actually used telephones. We
received competitive ETC money and built the wireless network there and today
80 percent of that population are wireless consumers.
The statistics Mr. Massey uses in his statements are simply not correct. Attached
is information from two government reports, Telephone Subscribership on American
Indian Reservations and Off-Reservation Trust Lands released by the Federal Communications Commission in May 2003 and the Telecommunications Challenges to
Assessing and Improving Telecommunications for Native Americans on Tribal Lands
released by the U.S. Government Accountability Office in January 2006. The reports, both based on the 2000 Census, provide the best neutral analysis of wireline
penetration rates on American Indian Reservations and Off-Reservation Trust
Lands. As stated, wireline penetration rates on the Pine Ridge Reservation were
greater than 75 percent for American Indian housing units by 2000. This level of
penetration is remarkable when one takes into account that the two poorest counties in the United States are part of the Pine Ridge Reservation.
Achieving this remarkable number was South Dakota-based Golden West Telecommunications, which serves over 48,000 customers in South Dakota (including
three reservations) with telephone, cable, high-speed Internet and other advanced
telecommunication services.
With regard to Mr. Masseys prc-filed testimony and oral testimony, Western
Wireless, now Alltel, did not deploy wireless service until November 2000 and did
not receive eligible telecommunications carrier status until October 2001. Given
this, it is evident that Mr. Masseys claim that telephone penetration rates on the
Pine Ridge Reservation improved from less than 30 percent to more than 80 percent
penetration is clearly inaccurate. It is not possible for them to claim any portion of
success given the timing of their service provision and the FCCs and GAOs docu-

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00128

Fmt 6601

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

125
mentation of wireline penetration rates. Alltel has stated these inaccurate numbers
in countless news articles and more than once in testimony before Congressional
panels.
I respectfully request that this letter along with its attachments be submitted to
the Committees hearing record for March 1, 2007 to ensure that inaccurate information does not continue to be presented regarding this matter. We recognize the
need for accurate information to reflect the reality of rural telecommunications so
that sound policy can secure the long-term success of the rural telecommunications
industry and for the Universal Service Fund. Thank you for your time and consideration on this matter.
Sincerely,
GEORGE STRANDELL,
General Manager and CEO,
Golden West Telecommunications.
cc: Hon. TED STEVENS,
Ranking Member,
Senate Commerce, Science, and Transportation Committee,
Washington, DC.
Hon. JOHN THUNE,
Member,
Senate Commerce, Science, and Transportation Committee,
U.S. Senate
Washington, DC.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00129

Fmt 6601

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

126
ATTACHMENTS
TELEPHONE SUBSCRIBERSHIP ON AMERICAN INDIAN RESERVATIONS AND OFF/
RESERVATION TRUST LANDS(DATA FROM 2000 DECENNIAL CENSUS)

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00130

Fmt 6601

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

301cw1.eps

127
GAO06189 TELECOMMUNICATIONS: CHALLENGES TO ASSESSING AND IMPROVING
TELECOMMUNICATIONS FOR NATIVE AMERICANS ON TRIBAL LANDS
*

RESPONSE

TO

WRITTEN QUESTIONS SUBMITTED BY HON. DANIEL K. INOUYE


HON. DEBORAH TAYLOR TATE

TO

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00131

Fmt 6601

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

301cw2.eps

Question 1. In 1997, the FCC adopted the principle that its Universal Service policies should be competitively neutral. In explaining this principle, the FCC concluded that competitive neutrality means that Universal Service support mechanisms and rules neither unfairly advantage nor disadvantage one provider over another, and neither unfairly favor nor disfavor one technology over another. But it
seems that people have different views as to how that principle should be applied,
particularly when it comes to providing support for different kinds of communications platforms. As members of the Joint Board, do you believe that this remains
a valuable principle, and how should it be applied to competition both among and
between communications platforms?
Answer. Competitive neutrality absolutely remains a valid and valuable principle.
Indeed, I believe that our goal should be a sustainable Universal Service system
that is, to the greatest extent possible, agnostic to the technology and platform supported. Our current rules, however, set support levels based on the costs incurred
by incumbent local exchange carriers (LEC) while allowing all other competitive eligible telecommunications carriers (CETCs)regardless of their coststo receive an
equal amount of support on a per line basis. While this means that all CETCs in
an area receive an equal amount of support per line, only the incumbent LECs sup-

128
port is actually cost based. Given the incredibly rapid growth of the CETC portion
of the high-cost fund, it is incumbent upon the Joint Board and the Commission to
consider whether these rules still make sense. Thus, the Joint Board is actively reviewing this equal support rule. See 47 C.F.R. 54.307. One potential alternative
to the equal support rule would be the use of reverse auctions to establish the
number of competitors and the level of support in an area, given a specific set of
service criteria. Reverse auctions present one way to identify the appropriate level
of ETC support in a market-based and competitively neutral manner.
Question 2. Section 254(c) of the Telecommunications Act of 1996 defines Universal Service as an evolving level of telecommunications services and also sets
forth criteria that the FCC considers when it decides which services qualify as supported services eligible for Universal Service support. At present, it is my understanding that the Universal Service Fund does not support broadband service. But
then, the question always arisesshould it? And if so, when? Do you think that
Universal Service should evolve to support broadband services, and if so, what
would trigger such a determination?
Answer. It is important to note that, in many instances, the Universal Service
Fund (USF) presently supports broadband services in an indirect manner. For example, carrier infrastructure investments funded through the USF frequently can
be upgraded to provide broadband at considerably reduced levels of expense and effort. In rural areas, schools and libraries connected to the Internet under the E-Rate
program often serve as anchor clients for advanced service providers that could
not otherwise economically provide broadband service to a community.
That being said, I believe that the USF should, and will, evolve to directly support
broadband services. Section 254(c) of the Act requires the Joint Board to consider
the evolving level of telecommunications services that should be supported, taking
into account advances in telecommunications and information technologies and services. In considering the evolution of supported services, the Act requires that we
consider the extent to which such telecommunications services(A) are essential
to education, public health, or public safety; (B) have, through the operation of market choices by customers, been subscribed to by a substantial majority of residential
customers; (C) are being deployed in public telecommunications networks by telecommunications carriers; and (D) are consistent with the public interest, convenience, and necessity. 47 U.S.C. 254(c). I believe that broadband may now, or soon
will, meet each of these standards.
Question 2a. Given that the law defines Universal Service as an evolving level of
telecommunications services and given that the FCC has classified cable modem
and DSL services as information services, would the Congress need to change the
statute to make broadband eligible for support?
Answer. Section 254(b) of the Act establishes access to advanced telecommunications and information services as a fundamental principle of universal service.
Section 254(c) of the Act requires the Commission to take into account advances
in telecommunications and information technologies and services in defining the
services that are supported by USF. The Commission may well, as a result, already
have the authority it needs to support advanced services that are not classified as
telecommunications services under the Act. Express clarification from Congress
would, however, eliminate any doubt.
Question 3. Currently the wireless eligible telecommunications carriers (ETCs) receive Universal Service support on a per customer basis based on the per line
costs of the wireline carrier in the same geographic area. This is sometimes called
the identical support rule and ensures that different communications platform providers receive the same amount of per line support. One criticism of the so-called
identical support rule for Universal Service is that it results in overly generous
support to wireless carriers because levels of support are not based on the per line
cost of providing wireless services. As a result, I have two questions
First, do you believe that Universal Service should support both wireline and
wireless services in rural America?
Answer. The Commissions current rules broadly support voice grade access to the
public-switched telephone network. Competitors, including both wireline and wireless carriers, bring a dazzling array of services to the rural areas of our country.
Indeed, wireless services have added a new dimension to connectivitymobility
that is very important to many consumers. Of course, as the steward of these consumer-derived funds, we must ensure that our policies are sustainable and will
allow new generations of Americans to have access to the latest generation of services so that our country is able to compete in the increasingly global economy. The
key is for consumers throughout the country have access to such services at just,
reasonable and affordable rates.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00132

Fmt 6601

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

129
Question 3a. Second, would it be possible to construct a model for wireless carriers
that would calculate support based on costs of wireless carriers, and what effect
might that have on the size of the fund?
Answer. It is my understanding from experts such as David Bodenhamer and Jim
Stegeman, each of whom testified at the en banc, that it would be possible to develop a wireless cost model. Such a model could provide some temporary reduction
in outlays from the fund. It is important to note, however, that the per line amount
of support provided to wireless CETCs is only one factor contributing to the incredible rate of growth of the CETC portion of the high-cost fund. Indeed, the fact that
we currently fund multiple networks in high-cost areasareas that require highcost support for even a single provider to servealso is a major ingredient in the
rapid growth of the high-cost fund.
Question 4. Commissioner Tate, it is my understanding that there are currently
two matters for decision pending before the Joint Board. One, referred by the FCC
in June 2004, examines what the rules should be governing the rural high-cost support mechanism. The other, referred by the FCC in November 2002, considers how
high-cost, Universal Service support should be calculated in competitive service
areas. Could you give the Committee a sense of when the Joint Board might make
a recommendation to the Commission on these issues? What steps must be taken
before any recommendation can be made?
Answer. The Joint Board is poised to act in the next several weeks to make a
short-term recommendation to stabilize the high-cost fund to the full Commission.
The Joint Boards recommendation on longer range solutions likely will take several
months longer. Joint Board recommendations are the product of an ongoing process
of negotiation and dialoguea process which currently is leading to significant forward progress.
Question 4a. Am I correct that the FCC must act on any recommendation made
by the Joint Board within 1 year?
Answer. Yes. Section 254(a)(2) of the Act states, the Commission shall complete
any proceeding to implement subsequent recommendations from any Joint Board on
Universal Service within 1 year after receiving such recommendations. I hope we
will be able to act more quickly.
RESPONSE

TO

WRITTEN QUESTIONS SUBMITTED BY HON. BILL NELSON


HON. DEBORAH TAYLOR TATE

TO

Question 1. There has been a lot of talk about reforming the USF contribution
assessment system. Much of this discussion has focused on moving toward a numbers-based system that would assess a per-line fee on all working telephone lines.
Do you believe that this could be implemented in a way that would not harm lowvolume and low-income telecommunications consumers?
Answer. Yes. Proponents of this change advocate that it will, among other things,
stabilize revenues, improve consumer understanding of the fees, and help to optimize use of our limited numbering resources. Others have voiced concerns that moving toward a numbers-based contribution assessment may negatively impact some
consumers, particularly low-volume and low-income consumers. While no proposal
is directly before me at this time, I believe that when the Commission does consider
reforming the USF contribution assessment system, we must carefully evaluate the
impact of each proposal on consumers. I remain open to ideas that will improve our
Universal Service contribution policies, but will insist that the solution we ultimately adopt be tailored to benefit, rather than burden, consumers.
Question 2. The concept of reverse auctions has been widely discussed as one solution to the problem of unchecked High-Cost Fund growth. How fast do you believe
a reverse auction program could be implemented? Why is it better than other approachessuch as study area caps or disaggregation? And, if implemented, what
sort of savings do you think reverse auctions would provide?
Answer. The amount of time it will take to implement a reverse auction program
will depend on the nature of the proposal. One proposal already in the record suggests a phased in approach that could take several years. Other proposals likely
could be implemented on a shorter timeline.
Reverse auctions present one way to identify the appropriate level of ETC support
in high-cost areas in a market-based and competitively neutral manner. This potentially is a technology neutral solution to one of the policy dilemmas we currently
face with our equal support rule, a policy that sets support levels based on the
costs incurred by incumbent local exchange carriers while allowing all other competitive eligible telecommunications carriersregardless of their coststo receive

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00133

Fmt 6601

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

130
an equal amount of support on a per line basis. Reverse auction rules also could
result in a reduction in the number of CETCs drawing high-cost support. The Joint
Board also is considering other policy proposals, such as disaggregation. The Joint
Board has not yet made a decision regarding which proposals it will recommend to
the Commission.
Question 3. Can reverse auctions be implemented in a manner that is truly competitively and technologically neutral? Wouldnt such a plan inevitably mandate
technology-based winners and losers?
Answer. I believe it is possible to create a reverse auctions system that is competitively and technologically neutral. Many commenters believe that this is one of the
significant benefits of utilizing reverse auctions. The Joint Board is cognizant of the
difficult changes that technological convergence is causing in the application of our
Universal Service policies and is working to make policy recommendations that recognize these marketplace changes in a manner that will promote access to advanced
telecommunications and information services at just, reasonable and affordable
rates throughout the Nation. Any recommendation made by the Joint Board will be
made on the basis of input from industry, state regulators, consumers, and other
stakeholders.
Question 4. Do you believe that the FCC currently has all the authority it needs
to implement a reverse auction process? What about authority to implement other
reforms (such as study area caps or disaggregation)?
Answer. Yesthe Commissions authority to implement Universal Service programs is broad. Any distribution mechanism would, however, necessarily have to adhere to the principles set forth in the Act. See 47 U.S.C. 254(b). While this is one
of the issues on which the Joint Board sought public comment last fall, I believe
that the Commission does have the authority necessary to implement a reverse auctions process. The Commission previously has instituted rules permissively allowing
disaggregation of support areas for certain purposes. See 47 C.F.R. 54.315.

RESPONSE

TO

WRITTEN QUESTIONS SUBMITTED BY HON. MARIA CANTWELL


HON. DEBORAH TAYLOR TATE

TO

Question 1. Commissioner Tate, under the current rules in place for the Universal
Service high-cost fund, can local exchange carriers obtain broadband equipment? If
so, under what circumstances can they obtain broadband equipment? Is there any
data regarding the extent to which local exchange carriers are obtaining broadband
equipment with Universal Service high-cost support funds?
Answer. Section 254(e) of the Act requires that, [a] carrier that receives [universal service] support shall use that support only for the provision, maintenance,
and upgrading of facilities and services for which the support is intended. Under
the high-cost fund, the Commission has permitted carriers to obtain support to be
used to upgrade loop facilities in a manner that permits the carrier to offer
broadband services in addition to voice service. See Federal-State Joint Board on
Universal Service, CC Docket No. 9645, Fourteenth Report and Order and TwentySecond Order on Reconsideration, Multi-Association Group (MAG) Plan for Regulation of Interstate Services of Non-Price Cap Incumbent Local Exchange Carriers and
Interexchange Carriers, CC Docket No. 00256, Report and Order, 16 FCC Rcd
11244, 1132023, paras. 194201 (2001) (Rural Task Force Order). The high-cost
fund does, therefore, indirectly support investment in broadband capable networks.
Other than this high-cost loop support, I am not aware of other high-cost support
mechanisms that directly support the acquisition of broadband equipment.
Question 2. Commissioner Tate, do you believe that legislative changes to the Universal Service Fund program should be completed prior to, concurrent with, or subsequent to any Commission action on intercarrier compensation? Do you see Universal Service reform and intercarrier compensation reform as linked or as separate
issues?
Answer. While guidance from Congress is always welcome, especially as we work
through the difficult legal and policy issues inherent to intercarrier compensation
reform, I believe that the Commission can take action in this area under the current
Act.
Intercarrier compensation reform is linked to Universal Service in some ways. For
example, the Missoula Plan would add significant payment obligations to the Universal Service Fund. Thus, while I believe that the two issues do not necessarily
have to be addressed simultaneously, reform of both systems must be complementary.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00134

Fmt 6601

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

131
Question 3. Commissioner Tate, in some rural parts of Washington State, there
are Wireless Internet Service Providers (WISP) that provide wireless phone service
and Internet access over the same device. Under the current rules could WISPs be
eligible to be an ETC as long as it provides wireless service?
Answer. Section 214(e) requires common carriers seeking ETC status to be designated an ETC by a state commission or the Federal Communications Commission.
See 47 U.S.C. 214(e). Wireline and wireless carriers designated as ETCs must offer
the telecommunications services or functions that are designated for USF support
by the Commission in Section 54.101 of its rules. See 47 U.S.C. 214(e)(1); 47 U.S.C.
254(c); 47 C.F.R. 54.101. ETCs also must file certifications that all support received will be used only for the provision, maintenance, and upgrading of facilities
and services for which the support is intended. See 47 U.S.C. 254(e); 47 C.F.R.
54.7.
RESPONSE

TO

WRITTEN QUESTIONS SUBMITTED BY HON. DANIEL K. INOUYE


HON. MICHAEL J. COPPS

TO

Question 1. In 1997, the FCC adopted the principle that its Universal Service policies should be competitively neutral. In explaining this principle, the FCC concluded that competitive neutrality means that Universal Service support mechanisms and rules neither unfairly advantage nor disadvantage one provider over another, and neither unfairly favor nor disfavor one technology over another. But it
seems that people have different views as to how that principle should be applied,
particularly when it comes to providing support for different kinds of communications platforms. As members of the Joint Board, do you believe that this remains
a valuable principle, and how should it be applied to competition both among and
between communications platforms?
Answer. I believe that competitive neutrality remains a valuable principle. Different types of technologies can benefit consumers and universal service. However,
I believe we need to take a closer look at how the system works today. We should
take into account the realities of the marketplace, the difficulties in achieving strict
competitive neutrality, and the differences in technology, including their costs, to
ensure that we dont unnecessarily favor one technology or company over another.
If we do that, I believe that consumers can benefit from multiple technologies while
the Universal Service system still supports them in a rational manner.
Question 2. Section 254(c) of the Telecommunications Act of 1996 defines Universal Service as an evolving level of telecommunications services and also sets
forth criteria that the FCC considers when it decides which services qualify as supported services eligible for Universal Service support. At present, it is my understanding that the Universal Service Fund does not support broadband service. But
then, the question always arisesshould it? Do you think that Universal Service
should evolve to support broadband services, and if so, what would trigger such a
determination?
Answer. I believe that the time has come to very explicitly include broadband as
part of our Universal Service system. I believe a good case can be made that the
Commission has statutory authority already to do this, but in light of FCC inaction
over the years, further guidance from Congress appears needed.
The Commission is charged with preserving and advancing universal service. That
means ensuring everyone, from the inner city to the most rural reaches of the country, has access to the wonders of communications. The challenge we face in meeting
this great objective is ensuring that our Universal Service mechanisms are sustainable. As more of our networks and communications migrate to broadband technology, I believe the key to sustainability lies in modernizing the Universal Service
system. That means having broadband both contribute to and receive support from
the Universal Service Fund.
Question 2a. Given that the law defines Universal Service as an evolving level of
telecommunications services and given that the FCC has classified cable modem
and DSL services as information services, would the Congress need to change the
statute to make broadband eligible for support?
Answer. When the Commission started down the road of reclassifying telecommunications services I was concerned that the Commission did not take the time
to think ahead to the possible intended and unintended consequences of our actions.
One serious source of concern was the real possibility that we would create impediments to bringing broadband to all of America. Nevertheless, I believe that Congress
provided the Commission with the statutory authority to make broadband eligible
for support when it told the Commission to base its Universal Service policies on
access to advanced telecommunications and information services. It may be the

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00135

Fmt 6601

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

132
case, however, that all of my colleagues do not support such a view. For this reason,
the Commission would benefit from additional Congressional guidance in this area.
Question 3. Currently the wireless eligible telecommunications carriers (ETCs) receive Universal Service support on a per customer basis based on the per line
costs of the wireline carrier in the same geographic area. This is sometimes called
the identical support rule and ensures that different communications platform providers receive the same amount of per line support. One criticism of the so-called
identical support rule for Universal Service is that it results in overly generous
support to wireless carriers because levels of support are not based on the per line
cost of providing wireless services. As a result, I have two questions
First, do you believe that Universal Service should support both wireline and
wireless services in rural America?
Answer. Yes. I believe that there is a place for wireless and wireline services in
our Universal Service system and we should treat them fairly.
Question 3a. Second, would it be possible to construct a model for wireless carriers
that would calculate support based on costs of wireless carriers, and what effect
might that have on the size of the fund?
Answer. It is clear to me that the costs of investing and maintaining wireless and
wireline infrastructure are inherently different. The Commissions current rules for
determining wireless eligible telecommunications carriers costs are both irrational
and costly as they are based on the wireline incumbent carriers costs. As I said at
the hearing, I believe that one of the things we can do to stabilize the Universal
Service Fund is eliminate the Commissions identical support rule. But to do so, we
need to have an alternative mechanism for calculating support based on wireless
carriers costs. Calculating these costs based on a model is certainly possible and
worth considering. At this time, it is difficult to know what the impact of a model
approach will have on the size of the Fund because the size of these costs will be
based on the models mechanics.

RESPONSE

TO

WRITTEN QUESTIONS SUBMITTED BY HON. BILL NELSON


HON. MICHAEL J. COPPS

TO

Question 1. There has been a lot of talk about reforming the USF contribution
assessment system. Much of this discussion has focused on moving toward a numbers-based system that would assess a per-line fee on all working telephone lines.
Do you believe that this could be implemented in a way that would not harm lowvolume and low-income telecommunications consumers?
Answer. My preference has leaned toward a revenue-based system, because it
makes intuitive sense that those who use the network more, pay more. But the current interstate revenue-based system may not be sustainable. The boundaries between local and long distance are eroding, while new Internet-based services are
growing. So the Commission may have to consider other approaches like a numbers
based approach to secure the future of universal service. But the way I see it, the
devil is in the details. Before supporting any such plan, I would need to understand
its impact on low-volume and low-income consumers. At the end of the day, whatever methodology we choose, I must be convinced that it benefits consumers.
Question 2. The concept of reverse auctions has been widely discussed as one solution to the problem of unchecked High-Cost Fund growth. How fast do you believe
a reverse auction program could be implemented? Why is it better than other approachessuch as study area caps or disaggregation? And, if implemented, what
sort of savings do you think reverse auctions would provide?
Answer. I am concerned about the impact of an auction-based Universal Service
system on rural areas in this country. So are many commenters on record at the
FCC. Congress charged the Commission with ensuring that consumers in all regions
of the Nation have access to comparable services at comparable rates. It is not yet
clear to me that an auction-based system would ensure adequate levels of support
and meet this Congressional objective. In addition, it appears that it would take
years before a reverse auction program could be implemented on a national basis
though it would be shorter to implement a pilot program as some have suggested.
There are many ideas other than reverse auctions that the Joint Board has before
it, including study area caps and disaggregation that would likely take a shorter
time to implement. Finally, without more detail on the types of reverse auctions to
be implemented it is difficult to determine what, if any savings, will be accomplished.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00136

Fmt 6601

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

133
Question 3. Can reverse auctions be implemented in a manner that is truly competitively and technologically neutral? Wouldnt such a plan inevitably mandate
technology-based winners and losers?
Answer. There are many different ways to implement a reverse auction. In fact,
both wireless carriers and wireline carriers have submitted proposals on how a reverse auction would work. Whether a reverse auction actually met the Commissions
policy of competitive neutrality would depend on the details of each proposal. However, it is not yet clear to me that an auction-based system that rewards the leastcost provider will guarantee comparable services at comparable rates, which is another core principle of universal service. When the Commission previously considered the use of auctions in 1997, it noted it is unlikely that there will be competition in a significant number of rural, insular, or high-cost areas in the near future.
Consequently, it is unlikely that competitive bidding mechanisms would be useful
in many areas in the near future. Before moving ahead here, it is imperative that
we understand what has changed since the Commission reached this conclusion. As
part of this analysis, we must consider whether any such proposal would be competitively and technology neutral.
Question 4. Do you believe that the FCC currently has all the authority it needs
to implement a reverse auction process? What about authority to implement other
reforms (such as study area caps or disaggregation)?
Answer. The Commission is charged with the preservation and advancement of
Universal Service based on the principles set forth in section 254(b), including ensuring that all Americans have quality services at reasonable rates, have access to
advanced telecommunications and information services, and have access to comparable service at comparable rates. To the extent that Universal Service proposals
concerning the distribution of funds, such as reverse auctions, study area caps and
disaggregation, are designed to comport with these principles, I believe that the
Commission has the authority to implement them. However, to the extent that Congress believes that a particular mechanism is inconsistent with these core principles, we would surely benefit from additional guidance.
RESPONSE

TO

WRITTEN QUESTIONS SUBMITTED BY HON. MARIA CANTWELL


HON. MICHAEL J. COPPS

TO

Question 1. Commissioner Copps, under the current rules in place for the Universal Service high-cost fund, can local exchange carriers obtain broadband equipment? If so, under what circumstances can they obtain broadband equipment? Is
there any data regarding the extent to which local exchange carriers are obtaining
broadband equipment with Universal Service high-cost support fund?
Answer. The High Cost program is already indirectly subsidizing broadband. Investments in telephone networks subsidized by the program end up subsidizing
broadband because most telephone equipment is capable of providing voice and data
services. Also, the Department of Agricultures Rural Utilities Service makes lowinterest loans to companies that invest in telephone networks capable of providing
broadband as well as voice telephone service. Many of those loans were made for
equipment that subsequently formed part of the cost basis for USF support. I am
not aware of data regarding how much high-cost support is used to support
broadband.
Question 2. Commissioner Copps, do you believe that legislative changes to the
Universal Service Fund program should be completed prior to, concurrent with, or
subsequent to any Commission action on intercarrier compensation? Do you see Universal Service reform and intercarrier compensation reform as linked or as separate
issues?
Answer. I believe that that one thing that could be done to stabilize the Fund is
to adjust the contribution rules to ensure that it is funded by intrastate and interstate revenues. With the boundaries between local and long distance eroding, and
the growth of any-distance calling plans, assessing only on interstate services is
growing more difficult over time. However, such a change would require action by
Congress. There does not appear to be a magic formula as to the timing of changes
to Universal Service and intercarrier compensation. However, any changes we make
to one program could require offsetting changes in the other. Thus any action must
be done in a comprehensive way.
Question 3. Commissioner Copps, in some rural parts of Washington State, there
are Wireless Internet Service Providers (WISP) that provide wireless phone service
and Internet access over the same device. Under the current rules could WISPs be
eligible to be an ETC as long as it provides wireless service?

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00137

Fmt 6601

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

134
Answer. The states have primary responsibility for designating telecommunications carriers as eligible telecommunications carriers (ETCs). It is my understanding that the FCC has not considered any carrier applications from Washington
State for designation as an ETC. In the case of Washington State, the Utilities and
Transportation Commission has the authority and has approved such applications.
Therefore, whether a wireless Internet service provider is eligible for ETC status
under Washington States rules is a decision for the state commission to determine.
RESPONSE

TO

WRITTEN QUESTIONS SUBMITTED BY HON. DANIEL K. INOUYE


LARRY S. LANDIS

TO

Question 1. In 1997, the FCC adopted the principle that its Universal Service policies should be competitively neutral. In explaining this principle, the FCC concluded that competitive neutrality means that Universal Service support mechanisms and rules neither unfairly advantage nor disadvantage one provider over another, and neither unfairly favor nor disfavor one technology over another. But it
seems that people have different views as to how that principle should be applied,
particularly when it comes to providing support for different kinds of communications platforms. As members of the Joint Board, do you believe that this remains
a valuable principle, and how should it be applied to competition both among and
between communications platforms?
Answer. I believe that competitive neutrality is a key concept, but so is seeing
that Universal Service funds are appropriately deployed, that legitimate needs are
met, and that accountability, efficiency, and performance are demanded. The focus
should be on the service provided to consumers not necessarily the companies or
technologies providing the service.
Seeking to establish a competitively neutral regime is an important principle,
but its application must be tempered by the specific legal requirements contained
in 47 U.S.C. 254(b)(3) and (b)(5) respectively. When it emerged, competitive neutrality seemed like it should be a driving focus and the only logical choice for policymakers. Logically enough, the issue has become achieving a competitively neutral
definition of competitive neutrality. Shifts in the market, including the pace of
changes occurring in technology, corporate consolidation, and corporate realignment
have been so great that this goal has become troublesome to operationalize with
specific policy reforms.
In designing a next-generation policy, we need to take some time not only to listen
to the lawyers but also to the marketers, since the former are paid to be close to
policymakers whereas marketers are paid to be close to the market.
The wireless industry has spoken justifiably, aggressively and articulately about
the achievements which have been made possible in an environment where there
is a light regulatory touch. And they have also pointed to those customers who have
cut the cord as evidence of wireless success story.
The reality however, is that the market still has significantly different expectations of the existing wireline and wireless technologies. The overwhelming majority
of customers have and use both technologies, but use them differently. Moreover,
customer expectations of wireless, while rising steadily, are still not competitively
neutral, if, for example, by competitive neutrality you mean number of dropped
calls, or other measures of quality of service.
A leading national wireless company has made fewest dropped calls the keystone of its marketing element, because that claim is meaningful to a large segment
of the wireless market. No wireline company would make such a claim because such
a claim would not be relevant to either the experience or the expectations of its customers. For equally obvious reasons, wireline companies choose not to compete in
the mobile convenience segment of the market.
Market-based issues aside, the requirements of reasonably comparable rates and
services as well as the pursuit of affordable rates and services are among the primary foci and drivers of our policies. Many predict we are moving toward the day
when there will be one converged, efficient network capable of provisioning multiple
layers of applications and services. Many companies have moved their business
models to this notion and means of operation, and to its companion marketing proposal . . . the triple or the quadruple play, with emphasis on expanding the
share of total communications wallet and driving both consolidation and partnering.
Question 2. Section 254(c) of the Telecommunications Act of 1996 defines Universal Service as an evolving level of telecommunications services and also sets
forth criteria that the FCC considers when it decides which services qualify as supported services eligible for Universal Service support. At present, it is my understanding that the Universal Service Fund does not support broadband service. But

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00138

Fmt 6601

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

135
then, the question always arisesshould it? And if so, when? Do you think that
Universal Service should evolve to support broadband services, and if so, what
would trigger such a determination?
Question 2a. Given that the law defines Universal Service as an evolving level of
telecommunications services and given that the FCC has classified cable modem
and DSL services as information services, would the Congress need to change the
statute to make broadband eligible for support?
Answer. As indicated by my colleagues in the March 1 hearing, there are specific
criteria set forth under 47 U.S.C. 254(c) which direct the Joint Board to recommend to the FCC, and for the FCC to establish, a definition of the telecommunications services which should be supported by Universal Service mechanisms.
Section 254(c) states that when adopting this list of telecommunications services,
the Joint Board and Commission shall consider whether the service is: (1) essential to education, public health, or public safety; (2) subscribed to by a substantial
majority of residential consumers; (3) being deployed by telecommunications carriers
in public telecommunications networks; and (4) consistent with the public interest,
convenience and necessity.
The Commission has concluded that each of these criteria must be considered,
but each not necessarily met, before a service may be included within the general
definition of universal service, should it be in the public interest.
In July of 2003, the FCC released its Definitions Order under CC Docket No. 96
45, upon which the Joint Board had made its recommendations. A part of that
Order was the consideration of advanced or high-speed services. The Commission
stated that it declined to expand the definition of supported services to include advanced or high-speed services at that time.
Although the Commission agreed with certain of those who filed comments in that
proceeding that broadband services were becoming increasingly important for consumers in all regions of the Nation, they also agreed with the Joint Board and the
majority of commenters that high-speed and advanced services currently [i.e., as of
2003] did not meet the Acts criteria for inclusion on the list of supported services.
Furthermore, the Commission went on to say that although telecommunications
carriers increasingly were deploying infrastructure capable of providing advanced
and high-speed services, the Commission agreed with the Joint Board and commenters that advanced services were not subscribed to by a substantial majority of
residential consumers. In fact, the Commissions own data showed that as of December 31, 2002, there were approximately 17.4 million high-speed lines serving residential and small business subscribers, which represented just 16 percent of all U.S.
households.
It is evident that broadband services are becoming increasingly pervasive and
moving in the direction of the substantial majority test, and it is also evident that
independent of the criteria set up in 47 U.S.C. 254(c), there is strong anecdotal
evidence to support the need for pervasive buildout.
It would be prudent for the FCC to refer this matter to the Joint Board for further
consideration, given that we may soon fulfill the substantial majority criterion.
The threshold legal question of whether or not advanced services, as they relate
to cable modems, DSL and other similarly situated services, would have to be reclassified from information services to telecommunications services for purposes of
USF support presents a potential legal quagmire for the FCC and the Joint Board.
Such reclassification could potentially reopen the door to litigation over jurisdictional and related issues which had been largely resolved.
It would seem to be less troublesome for Congress to pass a narrowly focused
amendment to the Telecommunications Act of 1996 allowing broadband services to
be supported, thereby investing the policy decision with statutory authority.
If a decision is made to move toward support of broadband services either through
the FCC or through Congress, it also becomes important to consider whether
broadband support should take the same general form as has High Cost support,
or whether that support should concentrate primarily or exclusively on the cost of
buildout. I believe the latter approach is preferable, and clearly more affordable.
The first question is whether and to what extent High Cost support is currently
advantaging the use of a second line for Internet access via dial-up, as opposed to
a single-line solution which rolls up POTS (or its VoIP surrogate) together with
high-speed access. An artificially depressed take rate created perversely through
a legacy technology subsidy could significantly impact the business plans of those
providers which are weighing broadband buildout and deciding where it is economical to implement it.
For other areas, implementation of a second tier incentive in the form of specific
tax breaks could prove sufficient to assure build out where the business model suggests that a positive return without incentives is unlikely or improbable.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00139

Fmt 6601

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

136
Finally, the Wyoming and Kentucky studies suggest there are remote areas (perhaps as few as 25 percent of total customers) where the cost of building out a single loop or equivalent may run as high as $10,000 or more, depending on circumstances and the technology involved. In such instances, a straight subsidy of
some sort is clearly required to produce the desired ubiquitous buildout.
However, the Wyoming study has also shown that significant intermodal distinctions exist among technologies. In that case, any provider (regardless of technology)
should be allowed to bid for the opportunity to build out to those highest cost customers, with the subsidy being awarded to the lowest bidder, but in no case should
a subsidy greater than the cost of the lowest price technology be granted.
Hypothetically, if the cost of buildout to a specific customer is projected at $2,500
for cable, $3,700 for wireless and $10,000 for a wireline provider, any provider
should be able to bid to serve that customer . . . but in no case should the winning
bidder receive more than the $2,500 representing the lowest cost provider.
Question 3. Currently the wireless eligible telecommunications carriers (ETCs) receive Universal Service support on a per customer basis based on the per line
costs of the wireline carrier in the same geographic area. This is sometimes called
the identical support rule and ensures that different communications platform providers receive the same amount of per line support. One criticism of the so-called
identical support rule for Universal Service is that it results in overly generous
support to wireless carriers because levels of support are not based on the per line
cost of providing wireless services. As a result, I have two questions
First, do you believe that Universal Service should support both wireline and
wireless services in rural America?
Second, would it be possible to construct a model for wireless carriers that would
calculate support based on costs of wireless carriers, and what effect might that
have on the size of the fund?
Answer. Yes, I believe that in truly high-cost rural areas, there is room both in
terms of public policy and funding to support both wireline and wireless services in
rural America. However, I have great difficulty accepting the notion that it is necessary for multiple wireless companies to receive what amount to multiple government subsidies in the process of bringing both wireless and wireline services to
truly rural areas.
In many if not most cases, the cities, towns and villages where the wireline providers locate their central offices and wire centers, are contestable. It is not unusual
in relatively small communities to find a wireline, wireless and cable provider competing for customers in those core areas.
I believe that the current system of equal support, while well-intentioned, does
little to achieve competitive neutrality. The identical support rule demonstrates the
unintended consequences which are produced when a strong desire to achieve competitive neutrality doesnt take into account differing cost structures.
Many in the industry readily admit in moments of candor that the identical support rule has become, in many instances, a means to game the system. Given the
current circumstances, State Commissions need to be more vigilant in their review
and approval process for CETC applicants, e.g., undertaking the sort of assessment
which the FCC contemplated in its 2005 ETC Order.1
Under the current structure, as CETCs, wireless carriers receive the same perline support as their wireline counterparts while in most cases their cost structure
is significantly less than that of the incumbent wireline carrier.
Also, as noted in Q. 1, carriers operate under a separate set of both policy and
service expectations. How can anyone realistically argue that wireline and wireless
companies are being treated in a competitively neutral manner when many wireless
companies and advocates admit privately that their costs are, in many cases, significantly lower that those of the ILECs?
At the same time, RLEC recipients of High Cost Line Support continue to receive
funding based on legacy investment and business decisions which may have been
made decades ago. It is arguable that this produces business decisions which resemble those of the Big 3 auto makers, with their legacy cost structure and legacy investments.
It is largely because of this identical support that we have seen first-hand the size
of the Fund grow exponentially over the last several years.
On an interim basis, as a preliminary threshold matter, I believe that CETCs
should receive support based on their own costs, not those of the incumbent wireline
carrier.
1 FCC

VerDate Nov 24 2008

14:09 Mar 21, 2012

Docket 9645, 0546 rel. March 17, 2005.

Jkt 073387

PO 00000

Frm 00140

Fmt 6601

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

137
I believe that a model could be constructed to capture a wireless carriers costs
as long as the relevant agency accounts for the type of territory that will be
serveddetermining, e.g., is it rural or urban, what type of terrain is it encompassing, farm land or mountainous? These types of considerations can be built into
a model in a much more efficient manner today than they were in previous
iterations of the current high-cost model.
As long as the companies seeking funding understand that they must justify that
funding with a cost analysis or model, I believe there is a strong incentivecurrently at $1 billion and doubling each year in recent yearsto provide such cost justification.
RESPONSE

TO

WRITTEN QUESTIONS SUBMITTED


LARRY S. LANDIS

BY

HON. BILL NELSON

TO

Question 1. There has been a lot of talk about reforming the USF contribution
assessment system. Much of this discussion has focused on moving toward a numbers-based system that would assess a per-line fee on all working telephone lines.
Do you believe that this could be implemented in a way that would not harm lowvolume and low-income telecommunications consumers?
Answer. Yes. I believe using a numbers-based methodology could be useful and
not harmful to low-volume or low-income consumers as long as the contributions
base is expanded to include wireless carriers, VoIP providers, and voice-grade
equivalents to capture special access and private lines, particularly for businesses.
The key to avoiding an undue burden for any segment of the population is to
spread the responsibility across all segments so that no one segment (i.e., low income users) is unduly burdened.
Currently, everyone who is connected to the system is receiving full value from
the system regardless of the price paid; if that were not the case, they would disconnect from the system. So we should not necessarily assume that just because
someone does not make a high volume of calls, that person is being harmed by the
price per call paid. That person may place a higher value on each call, or simply
on the ability to access the network at will, than does the higher volume user.
In the event Congress should determine that additional steps should be taken to
avoid burdening low-income consumers, there are multiple options available, including increasing Lifeline support or indexing to income level, but such a move should
be based on appropriate and totally objective 3rd party data to make certain the
focus is squarely on the target population.
Question 2. The concept of reverse auctions has been widely discussed as one solution to the problem of unchecked High-Cost Fund growth. How fast do you believe
a reverse auction program could be implemented? Why is it better than other approachessuch as study area caps or disaggregation? And, if implemented, what
sort of savings do you think reverse auctions would provide?
Answer. I believe that if implemented nationwide, a reverse auction system could
be functional within two to 3 years. I would personally prefer an approach in which
reverse auctions were first tested, perhaps as a means of identifying providers in
unserved areas or of selecting a winner or winners (depending on the model
adopted) in a representative group of states, before being implemented nationally.
Many key questions must be addressed, as Im sure the members of the Committee
are well aware.
Every aspect of the design of a reverse auction needs to be carefully considered,
especially including who wins, which is directly related to the question of whether
the design is to be winner takes all, winner takes more [as proposed in at least
one wireless CETC auction model in response to the Request for Comment], or
whether there is one winner for each sectori.e., wireline and wireless.
If the design which is ultimately selected were to be one winner takes all, there
are numerous issues to be resolved in the event the incumbent local exchange
(wireline) carrier is not successful in the auction process, including what happens
to that incumbents network and overall presence in the market if the wireline incumbent is unsuccessful and does not win the auction.
Since most wireless providers are still dependent on wireline incumbents networks for transport, the question of what constitutes fair compensation for continued use of portions of the network of a losing incumbent by a winning CETC
is a critical issue.
Further, we also need to be prepared to offer a transition mechanism for incumbents that might lose in an auction setting, since their business plans are premised on the current USF disbursement system rather than on a significantly different mechanism.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00141

Fmt 6601

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

138
In our current proceeding we are examining how, on a broad basis, we deal with
the possibility of partial or full re-monopolization of the marketplace in certain regions. This threshold policy question is important because some rural and insular
areas of the United States may not be able to support more than one carrier, yet
under the current USF structure, multiple carriers are funded. This is a political
and economic reality, yet we are very mindful that this system cannot be unwound
overnight, assuming a consensus emerges that implementation of a reverse auction
approach is a worthy goal.
That is why the answers are dependent upon the implementation rulesi.e., the
design and structure of the auction[s] if such an approach were to be adoptedand
without such details being addressed, it is almost impossible to quantify and assess
what possible savings or costs would be produced.
Interim measures, including a cap on funding, are essential in order to stem the
accelerating growth spiral of the Fund in the short term. As my colleague Billy Jack
Gregg said in response to a question at the March 1 hearing, that interim cap must
be applied where the problem exists . . . where the growth is occurring. Like Willie
Sutton in Director Greggs response, we need to go where the money is, and apply
the temporary cap there.
Put another way, an EMT responding to a serious accident does not apply a tourniquet to a victims leg if that accident victim is hemorrhaging from the arm. And
this analogy is appropriate in more ways than one, because a proposed interim cap,
like a tourniquet, is only intended to be a temporary measure to address an acute
need until the patient can be fully triaged and comprehensively treated.
This step would allow the Joint Board sufficient time to address longer-term
issues without leaving the Fund in jeopardy of implosion because it cannot sustain
itself. It also has the advantage of being an admittedly imperfect remedy, dramatically increasing the likelihood that neither the Joint Board nor the FCC will find
it a necessary and sufficient solution and thereby make it possible to declare victory.
Disaggregation is not a new concept for the FCCs consideration. In the Rural
Task Force Order of 2001, the FCC recommended disaggregation for rural carriers;
however, only a small minority of all rural carriers took advantage of this opportunity. I believe that disaggregation is essential and not incompatible with a properly-designed reverse auction solution or other alternative for USF reform, including
a models-based solution, provided anti-gaming protections are built in.
Question 3. Can reverse auctions be implemented in a manner that is truly competitively and technologically neutral? Wouldnt such a plan inevitably mandate
technology-based winners and losers?
Answer. Yes, a reverse auction approach can be adopted which is both competitively and technologically neutral. As I have advocated previously, a reverse auction
may well need to be linked to other reforms such as disaggregation to assure both
neutrality and compliance with legislative intent.
Taken as a whole, the current framework, well-intentioned as it was, is all about
choosing winners and losers. In too many cases, providers are not being held to account or expected to appropriately steward the funds which they receive. A relative
handful of states are according a virtual free pass to USF funding through their failure to implement the voluntary guidelines for screening CETC applicants promulgated by the FCC in its March 17, 2005 ETC Order.1
As long as these circumstances exist, those of us who shape public policywhether
serving in Congress, on the FCC, or as a member of a State Commissionare choosing winners, even if that is usually as an act of omission rather than of commission.
When we do so, we are also choosing losers: the American people, in the form of
higher-than-necessary USF levies placed upon ratepayers.
Virtually any form of reverse auction which has been discussed in conjunction
with the Joint Boards current proceeding will produce losers, by design. The
whole purpose of a reverse auction model is to derive the greatest value for the least
possible investment of high-cost dollars for the customer. In some circumstances
one technology may advantage its user over the differing technologies of other providers competing to serve in the same geographic area. But in a reverse auction,
each provider is free to determine how little s/he is willing to accept in return for
the franchise to serve an area. If s/he is willing to pay more dearly by accepting
a lower level of compensation than the competitionregardless of cost differentials
which may exist intermodallythen s/he will be the winner. Thus the winner
is the bidder who brings the greatest value for the least cost to the customer. The
question is whether it would be deemed politically acceptable.
1 FCC

VerDate Nov 24 2008

14:09 Mar 21, 2012

Docket 9645, 0546 rel. March 17, 2005.

Jkt 073387

PO 00000

Frm 00142

Fmt 6601

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

139
The winners/losers issue depends on vendors definitions of the value of subsidy.
A higher cost vendor may elect to receive a lower margin than his/her competitors,
in order to retain or gain the incremental revenue produced through high-cost support. So the vendors themselves determine what is fair, by the full value they
place on winning.
When you let the bidders determine the value of the subsidy they are eligible to
receive as the successful bidder or bidders, then by definition the winner receives
full value, and all losers are losers because they set a higher value than the market (through the reverse auction mechanism) was willing to attach to provision of
service.
RESPONSE

TO

WRITTEN QUESTIONS SUBMITTED BY HON. DANIEL K. INOUYE


HON. JOHN DOWNES BURKE

TO

Question 1. In 1997, the FCC adopted the principle that its Universal Service policies should be competitively neutral. In explaining this principle, the FCC concluded that competitive neutrality means that Universal Service support mechanisms and rules neither unfairly advantage nor disadvantage one provider over another, and neither unfairly favor nor disfavor one technology over another. But it
seems that people have different views as to how that principle should be applied,
particularly when it comes to providing support for different kinds of communications platforms. As members of the Joint Board, do you believe that this remains
a valuable principle, and how should it be applied to competition both among and
between communications platforms?
Answer. Competitive neutrality seemed an appropriate and important principle
when the FCC and the Joint Board were first trying to implement the 1996 Act.
At that time the overwhelming objective of Federal policy was to open the local exchange network to competition, and many new policies were aimed at giving a boost
to the infant CLEC industry. But Universal Service was also an important goal
under the Act. Competitive neutrality should not be an obstacle to the primary goal
of preserving and advancing universal service. Neutrality does not provide a sufficient basis for a subsidy that does not demonstrably advance Universal Service
goals, particularly when it inflates the contributions required from telephone subscribers.
The Identical Support
Competitive neutrality is often today considered synonymous with the Identical
Support Rule (ISR). Under ISR, a Competitive Eligible Telecommunications Carrier (CETC) receives per-line support equal to that given to the Incumbent Local
Exchange Carriers (ILEC) serving a customer. If ILEC A and an CETC B have
customers with billing addresses in the same ILEC wire center, they receive equal
per-line support.
The ISR, although motivated by competitive neutrality, is not neutral. The ISR
allows that carriers A and B both get support based on carrier As costs. A competitively neutral rule would, for example, award support to each based on its own cost;
or it might award support to both based upon some third factor not dependent upon
eithers network.
Nor does the ISR produce competitively neutral results. While the support
amounts are the same for A and B, they may have vastly different cost structures.
Most CETCs are wireless carriers. Wireless technology finds classical wire center
boundaries largely irrelevant. Wireless carrier costs and deployment are significantly affected by factors, such as topography, that are less important for wireline
carriers. Moreover, for many rural customers, their address is largely irrelevant because they cannot receive wireless service at home, but subscribe because they travel. There is no reason to believe that a support dollar given to a CETC under the
ISR will produce results anything like the results of giving that dollar to an ILEC.
A second major problem with the ISR is that it, when combined with the legacy
procedures for calculating ILEC support, has generated enormous Fund growth.
CETCs support has been growing at an annual rate of 101 percent since 2002.
CETCs received almost $1 billion in 2006. Because wireless carriers are now applying for CETC status in droves, CETCs are on track to receive over $1.5 billion in
2007.1
The ISR did not contemplate that a household would retain its landline and add
three or four supported wireless phones as well. Under the ISR, all of these phones
can receive an equal subsidy.
1 AT&T

VerDate Nov 24 2008

14:09 Mar 21, 2012

ex parte filing in Docket No. 9645 on March 22, 2007.

Jkt 073387

PO 00000

Frm 00143

Fmt 6601

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

140
There is another and more complex reason for the increase. ISR works in tandem
with legacy support mechanism of ILECs.2 The following table illustrates how competition could increase total support in a high-cost area more than tenfold, even
without the total number of lines increasing.3
Network
Operation
Cost
(000s)

No competition
Slight
competition
Intense
competition

ILEC
lines

ILEC
per-line
cost

ILEC
Support
per line

ILEC
Support
(000s)

$53

$532

CETC
Lines

CETC
Support
(000s)

Total
Support
(000s)

$532

$1,000

10,000

$100

900

8,000

113

63

502

2,000

$125

627

800

1,000

800

585

585

9,000

5,267

5,852

Commonly, CETC access lines increase faster than ILEC lines decrease. Modifying the preceding analysis to reflect that fact would only increase CETC support
more rapidly.
The combined effect of these ISR and legacy ILEC mechanisms has been to subsidize the construction of second, third and fourth networks in high-cost areas where
it was historically difficult to finance construction of the original network. This
might be a good result in the narrow case of a wireless network extending service
to a previously unserved rural area, but is not a sound general policy. There is no
indication that the USF was designed, even in part, to subsidize robust competition
in hard-to-serve areas.
A third major problem with the ISR is that it generally provides too much or too
little support to CETCs. Even if one assumes that there is sufficient support for
ILECs,4 the ISR generally will produce more-than-sufficient or less-than-sufficient
support for CETCs.
As the table above shows, support to CETCs can easily be more than sufficient.
In a market with intense competition, a CETC with a high market share can receive
many times the per-line support ($585) that the incumbent received before competitors arrived ($53). This is particularly incongruous if the CETC has superior technology that provides telecommunications services more efficiently than the legacy
technology.
Support to CETCs can also be less than sufficient to support a network capable
of serving all customers in the service area. Two wireless CETCs might share a
market, for example, and might have similar facilities and receive equal per-line
support payments. Yet if one CETC has an 50 percent market share and the second
CETC has a 1 percent market share, the first will receive 50 times as much support
as the second. This is no recipe for maintaining continued service by the second
CETC. On the contrary, this is a potentially unstable condition in which Federal
support might be used by larger carriers to drive out smaller competitors.
A fourth major problem with the ISR is that it awards support without any clear
objective or meaningful performance expectation. The great majority of CETC
money actually goes to wireless carriers, but this has little demonstrable relation
to Universal Service goals.
Wireless carriers generally offer nationwide rates. Therefore, although support to
a wireless carrier may promote greater service availability, it does not have an effect on whether a customer living in a high-cost area receives wireless service at
affordable and comparable rates.
Much of the money transferred to CETCs under the ISR is based on the IAS
and ICLS programs. This funding for ILECs was historically derived from specific
access rate reduction decisions by the FCC. The connection to CETCs is tenuous.
2 Legacy support mechanisms for ILECs are intended to provide sufficient support for the
ILECs to operate a wireline network. When an ILEC has fewer lines supporting its network
that generally increases per-line cost, and that increases per-line support for the ILECs remaining customers. The EPR then equally increases per-line support to CETCs. Total support thereby increases rapidly as more customers are served by CETCs.
3 The support mechanism illustrated here is to provide support equal to 76 percent of the difference between per-line cost per month and $30 per month. This is most similar to the support
mechanism current used for nonrural carriers. The intense competition example here assumes
that the ILEC loses 90 percent of its original lines. This is an extreme case chosen to illustrate
the point, but it is not totally implausible given the inroads now being made by wireless and
VoIP, and the impending widespread availability of VoIP over cable systems.
4 The Tenth Circuit Court of Appeals found in 2005 that the FCC had not demonstrably provided sufficient support to customers of so-called nonrural carriers. Qwest v. FCC, 398 F.3d
1222 (10th Cir. 2005).

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00144

Fmt 6601

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

141
In other words, CETCs today are receiving millions of dollars in support because
the local ILEC once had high interstate access rates. For the ILECs, this money
may once have had a connection to Universal Service objectives. Porting it over to
CETCs does not significantly advance any Universal Service objectives.
Alternatives
The arrival of real competition in high-cost rural areas forces us to face several
unpleasant alternatives. If we keep the ISR, the Fund will continue to grow exponentially, and we will pay more to CETCs than they need to provide service.
On the other hand, abandoning the ISR will require us to grapple with some difficult problems. The most immediate would be to determine how many networks
merit support in a given area, and how they should be selected. This could conceivably be done by auctions, but auctions have myriad administrative difficulties, and
it is not clear these can all be solved.
Second, we would need to decide whether the inherent differences among networks should affect support. Different networks have different cost structures and
present different subsidy considerations. Wireless uses different types of facilities
with different kinds of propagation characteristics, and this certainly leads to a cost
structure that is much different than ILECs. Moreover, wireless carriers often build
facilities to serve customers who have billing addresses tens or hundreds of miles
away. Wireless carriers also have different kinds of revenues and costs from intercarrier transactions. All of this could be relevant to a support mechanism for wireless carriers.
Another possibility is to cap the total support offered in a study area or state.
This has significant risks also. Capped support may not be sufficient for ILECs to
keep their retail rates affordable and reasonably comparable. With a cap, there is
a risk that all networks would fail or, more likely, constrain service to the lowest
cost areas. Failure of the ILEC would be particularly problematic where a CETC
depends (as do most wireless carriers) on the ILEC for network transport functions.
Finally, one could adopt separate wireline and wireless Universal Service systems.
Such a system would be able to acknowledge the differences between the technologies, derive an appropriate business model for each that leads to a support
amount, and thereby provide sufficient support to meet the statutory objectives in
Section 254.
Conclusion
My primary conclusion is that the principle of competitive neutrality should be
made secondary to other, more important, principles, such as:
Universal Service support payments should produce a demonstrable benefit to
consumers, either in the form of reduced rates or increased availability.
Funding should be sufficient to ensure that customers everywhere have access
to at least one telecommunications service that provides acceptable service at
comparable rates. That should include broadband service, at a specific date in
the future.
Public funds should not be provided automatically to every network that is constructed with private capital. Subsidies should go to only a limited number of
networks in high-cost areas. For discussion purposes, I would suggest that funding be available in any area to only one wireline and one wireless network. This
is not intended to limit in any way the uses of private capital or to limit competition among privately financed networks.
Universal Service policy can legitimately differentiate among competing telecommunications technologies. Funding should impose an obligation to meet
minimum standards, even if this would effectively disqualify a particular technology.
Universal Service support should be based upon the reasonable financial needs
of the supported carrier. This requires consideration of revenues available from
all sources, including intercarrier revenues and all subscriber revenues from
regulated and nonregulated activities.
My second conclusion is that the ISR should be abandoned. No carrier should receive support based upon another carriers costs and revenues. Support to wireless
carriers could conceivably be based upon a wireless cost model, but these carriers
may also need to submit actual cost information and actual facility location information as a prerequisite to support.
Question 2. Section 254(c) of the Telecommunications Act of 1996 defines Universal Service as an evolving level of telecommunications services and also sets
forth criteria that the FCC considers when it decides which services qualify as sup-

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00145

Fmt 6601

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

142
ported services eligible for Universal Service support. At present, it is my understanding that the Universal Service Fund does not support broadband service. But
then, the question always arisesshould it? And if so, when? Do you think that
Universal Service should evolve to support broadband services, and if so, what
would trigger such a determination? Given that the law defines Universal Service
as an evolving level of telecommunications services and given that the FCC has
classified cable modem and DSL services as information services, would the Congress need to change the statute to make broadband eligible for support?
Answer. Yes, Universal Service should evolve to support broadband services. That
decision should be made now, and reasonable target dates should be set for compliance. For example, it might be reasonable to set a target that 95 percent of the
American public would have access to a broadband service by 2010. In this context
access would mean that a person could purchase broadband at his or her residence
or place of business from one or more sources at a rate that is reasonably comparable to urban rates. Broadband should be defined in a way that encourages maturation of the network, but that does not disqualify services that are now widely
subscribed to.
Congress should amend section 254 of the Act. Regardless of whether it explicitly
defines broadband as an eligible service, it should amend subsections 254(c) and
254(e).
Subsection 254(c)
Subsection 254(c) envisions a binary decision on telecommunications services;
services are either included or not included. If broadband were defined as an eligible service, then three consequences follow: (1) broadband would become part of the
minimum standard for eligibility as an Eligible Telecommunications Carrier
(ETC); (2) broadband costs may be considered when calculating support; and (3)
ETCs may spend Universal Service support to maintain broadband facilities. In
other words, the current statute requires that each particular service be both mandatory and permitted, or neither.
Some existing ETCs do not offer broadband to all their customers. Therefore if
broadband is added to the list under section 254(c), some existing ETCs might be
disqualified unless they could offer broadband to all their customers. In my opinion,
such mass disqualification would be undesirable, but it is not clear how to avoid this
result if broadband is added to the list of supported services. This tension between
mandatory and permissive services makes it difficult for the FCC to add services
to the list. Although the express intent of section 254 was for Universal Service
standards to evolve over time, the structure of section 254(c) makes such evolution
a very high stakes process with possibly punitive results.
I do not want to imply that the statute has been an absolute barrier to progress.
Universal service payments have actually supported the development of broadband
networks in some areas. Notably, the High Cost Loop (HCL) program, the FCCs
largest single high-cost program,5 supports the loop cost of hundreds of smaller
so-called rural carriers. HCL support is calculated based on the carriers investment level in its loop facilities, the wires and distributed platforms that are outside central offices. Many rural carriers have built broadband networks capable of
providing DSL services, and some have even built fiber networks capable of delivering video. In most cases these investments have generated HCL support.
Not only does the HCL program support broadband buildout, it seems to have developed a preference for broadband. The HCL program is capped. Many rural carriers increased their per-line investment, and the differential effect has been to
draw support away from carriers that merely provide voice service, while adding
support to carriers that have broadband-capable and even video-capable networks.
Thus, although broadband has not yet become a supported service, one major FCC
program is currently providing de facto support for broadband. This result was explained through the Joint Boards policy of avoiding barriers to broadband.6 While
5 The

HCL programs annualized cost is $1.4 billion.


policy was suggested in 2000 by the Rural Task Force, which said that there should
be no barriers to advanced services. See Federal-State Joint Board on Universal Service, CC
Docket No. 9645, Rural Task Force Recommendation To The Federal-State Joint Board On Universal Service, released Sept. 29, 2000, at 22 (policy should incorporate the following principles:
a. Universal Service funding should support plant that can, either as built or with the addition
of plant elements, when available, provide access to advanced services. State commissions could
facilitate this infrastructure evolution and may make an exception for carriers with functional
but non-complying facilities. b. Telecommunications carriers should be encouraged by regulatory
measures to remove infrastructure barriers relating to access to advanced services. c. The Federal Universal Service support fund should be sized so that it presents no barriers to investment
in plant needed to provide access to advanced services. Specifically, to remain sufficient under
6 This

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00146

Fmt 6601

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

143
this was a creative way to advance broadband deployment, it has been controversial,
and it has not been applied to all carriers equally. In most states, nearly all of the
customers of rural carriers have access to advanced broadband networks; but it
is also common to learn that their rural neighbors who happen to be served by larger carriers cannot get DSL.
Congress should consider amending section 254 in a way that authorizes support
for broadband, but that does not unintentionally disqualify existing ETCs. The chosen path should provide support in rural areas without regard to the size of the carrier that happens to serve the area. One step in the right direction would be to
amend section 254(c) to allow some services to be supportable without also being
mandatory.
Subsection 254(e)
Congress might also clarify the statutory injunction in section 254(e). This statute
limits the uses of support only for the provision, maintenance, and upgrading of
facilities and services for which the support is intended. Several ambiguities arise
under this language. Incumbent LECs sometimes argue that this language requires
no more than that they continue to provide minimally adequate service over their
existing facilities. Since all Federal high-cost support becomes revenue to these carriers, and since they provide the minimally acceptable level of service, this standard
in some cases does not have much effect on either services or rates.
Under FCC guidance, ETCs must today make annual reports to the FCC.7 Those
reports must include a report on progress under the carriers five-year service quality improvement plan. 8 However, nothing in the current rules requires network improvement plans to include broadband by any date-certain. The reports must also
include the number of requests for service from potential customers . . . that were
unfulfilled during the past year and how the carrier attempted to provide service
to those potential customers. 9
More rigorous approaches are possible. For example, Verizon-Vermonts Modelbased high-cost support increased significantly in 2000. That increase has been distributed as explicit credits on monthly customer bills.10
If Congress wishes to achieve more significant or more demonstrable results from
Universal Service support, it might clarify subsection 254(e). One option would be
to mandate that all high-cost funds appear as explicit credits on customer bills.11
Alternatively, if Congress wishes to allow carriers to continue treating Universal
Service support as carrier revenue, it might directly mandate that carriers adopt
service quality improvement plans and further mandate that those plans call for
broadband by a specific date.
Information Services
As the last part of the question suggests, recent FCC decisions narrowing the definition of telecommunications services may have created a barrier to providing Universal Service support to broadband.
Some broadband services still are telecommunications services,12 but the FCC has
declared a wide range of retail broadband services to be information services, including cable modem service 13 and facilities-based wireline broadband Internet access service (DSL).14 Moreover, the FCC has repeatedly stated that when a service
is an information service it cannot also be a telecommunications service.
the 1996 Act, the Fund should be sized so that investment in rural infrastructure will be permitted to grow.)
7 See 47 C.F.R. 54.209.
8 See 47 C.F.R. 54.209(a)(1).
9 See 47 C.F.R. 54.209(a)(3).
10 The current credit for residential customers is $1.41. Higher credits are given to business
customers, because they have larger bills.
11 Carriers are currently explicitly permitted to show Universal Service contributions as explicit charges, and all or nearly all do so.
12 For example, special access circuits are broadband services, but they do not necessarily connect to the Internet.
13 Inquiry Concerning High-Speed Access to the Internet Over Cable and Other Facilities, Internet Over Cable Declaratory Ruling, Appropriate Regulatory Treatment for Broadband Access to
the Internet Over Cable Facilities, GN Docket No. 00185 & CS Docket No. 0252, Declaratory
Ruling and Notice of Proposed Rulemaking, 17 FCC Rcd 4798 (2002) (Cable Modem Declaratory
Ruling and NPRM); affd National Cable & Telecommunications Assn v. Brand X Internet Services, 125 S. Ct. 2688 (2005).
14 Appropriate Framework For Broadband Access To The Internet Over Wireline Facilities, CC
Docket No. 0233, Report and Order and Notice of Proposed Rulemaking, released Sept. 23,
2005, 20 FCC Rcd. 14,853.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00147

Fmt 6601

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

144
Various provisions of 47 U.S.C. 254 suggest that Federal support may be provided only to support telecommunications services. For example, subsection (c)(1)
says that [u]niversal service is an evolving level of telecommunications services.
A more specific passage in that same paragraph states:
The Joint Board in recommending, and the Commission in establishing, the definition of the services that are supported by Federal Universal Service support
mechanisms shall consider the extent to which such telecommunications services. . . .15
This implies that only telecommunications services may be included in the definition of services that are supported by Federal Universal Service support mechanisms. The argument is only strengthened by subdivision (c)(3) which allows the
schools and libraries and healthcare programs to support additional services not
on the official list.
On the contrary, it is also clear that section 254 establishes an overall goal of promoting access to advanced services. This is evident in subdivision (b)(2) which sets
the goal of providing Access to advanced telecommunications and information services in all regions of the Nation. However, such a general goal may not be sufficient
to override specific contrary terms in the operational parts of section 254.
On balance, I believe that under the existing statute there are serious questions
about:
1. whether Federal funds may be used to support services that are not on the
list of supported services under section 254(c)(1); and
2. whether an information service can be a supported service.
I recommend that Congress clarify subsection 254(c) on these points.
Question 3. Currently the wireless eligible telecommunications carriers (ETCs) receive Universal Service support on a per customer basis based on the per line
costs of the wireline carrier in the same geographic area. This is sometimes called
the identical support rule and ensures that different communications platform providers receive the same amount of per line support. One criticism of the so-called
identical support rule for Universal Service is that it results in overly generous
support to wireless carriers because levels of support are not based on the per line
cost of providing wireless services. As a result, I have two questions
First, do you believe that Universal Service should support both wireline and
wireless services in rural America?
Second, would it be possible to construct a model for wireless carriers that
would calculate support based on costs of wireless carriers, and what effect
might that have on the size of the fund?
Answer.
Wireline and Wireless
Yes, I do believe that Universal Service should support both wireline and wireless
services in rural America. However, as I explained above in some detail I have serious reservations about the Identical Support Rule.
We should seriously consider supporting wireless under a separate program. This
would allow the separate programs to be designed more sensibly. They could reflect
differences in signal propagation characteristics, differences in the extent of existing
deployment, different deployment and service goals, and possibly by different expectations about how funds will be used and accounted for.
I would encourage the Congress to authorize matching grants as a way to increase
deployment of both wireless and wireline broadband technologies. While these technologies can create significant revenue streams once facilities have been built, the
initial construction costs are daunting, particularly in areas of low density and in
areas where rugged terrain limits the propagation of wireless signals.
In my state, we face very real limits in current broadband deployment; but we
are working hard to improve the situation. I dont think Vermont is unique in this
regard. States can have very detailed and relevant knowledge about where
broadband improvements are needed. Particularly if Federal funds are matched,
Congress can be sure they will be spent wisely.
Wireless Cost Model
Yes, it is possible to construct a model of wireless costs, and that work is largely
complete. A commercially available model was used, for example, in the recent Wyo15 See

VerDate Nov 24 2008

14:09 Mar 21, 2012

47 U.S.C. 254(c).

Jkt 073387

PO 00000

Frm 00148

Fmt 6601

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

145
ming project that Commissioner Landis described in his March 1 oral testimony.16
There might be issues adapting it for FCC purposes, such as making all of the inputs public, but most of the technical challenges have been solved.
However, that is only a partial answer because a cost model only calculates costs.
One also needs a support model to calculate support. Support models require additional data inputs and policy decisions. Support models also present the most difficult policy challenges, because they confront most directly the tension between
competition and universal service.
One difficult issue for support models is carrier revenues. The common goal of
support programs is to manage the payments that the carrier ultimately demands
from its subscriber. Support is adjusted to keep these payments within limits, using
standards such as affordable or reasonably comparable. In reality, those customer payments are affected by the carriers entire business model, and that certainly includes payments made to and revenues received from other carriers. Carriers now produce many kinds of revenues, only some of which are regulated in
the classical sense. Although this is a complex policy area, current FCC support
mechanisms could be improved to make these revenue assumptions more explicit
and more comprehensive. Notably, we should carefully consider whether to include
revenue from unregulated services.
Another difficult issue is the location of the service. Support models typically associate support with particular areas, and they locate customers by their billing addresses. For wireline service, this practice makes sense because wireline customers
typically receive and use the service at their billing location. That is not true for
wireless, however. A wireless customers billing address can have little relationship
to service; some wireless customers cannot even get a signal at home. Moreover,
wireless facilities are often constructed primarily to serve passing customers who
are billed in other places. Consider a cell tower located next to a remote stretch of
interstate highway. The number of customers who have a nearby billing address
bears no relation to the reason that tower was built, and it would be a mistake to
assume that the tower is serving only those customers who have local billing addresses.
The most difficult issue for both cost models and support models is how to account
for multiple networks. Consider wireline carrier A that provides retail telephone
service and also provides special access. Consider also wireless carrier B that uses
As special access circuits to connect its cell sites.
First, consider the complications for the cost model. A realistic estimate of Bs
costs requires one to know how much B must pay A for those special access circuits.
Furthermore, A and B may share some facilities.17
For the support model the issues are even more complex, and they are fraught
with policy judgments about whether support should be used to maintain multiple
networks. Carrier As support might be reduced, for example, to the extent it draws
(or ought to draw) special access revenue from B. Conversely, if we expect B to continue providing service, it might be necessary to increase As support as B captures
more of As retail customers. This is exactly contrary to the original expectations
of the Joint Board about how competition would affect support, which assumed that
support would move away from A when B captured As customer.
In short, even if one has already developed a good wireless cost model, calculating
support still requires resolution of several significant policy issues. Therefore, the
availability of a wireless cost model is only the first of several difficult steps in developing a support mechanism.
The current size of the high-cost fund, $4.3 billion, is probably sufficient to maintain a quality wireline and a quality wireless network. However, there are major
tasks ahead. We should establish universal or near-universal wireless coverage and
broadband coverage, and we should reduce existing inequities that treat some rural
customers much better than others. Solving these problems will require us to make
a difficult choice between either allowing the Fund to grow still larger or reallocating existing support by employing a new and more rational allocation system.
RESPONSE

TO

WRITTEN QUESTIONS SUBMITTED BY HON. BILL NELSON


HON. JOHN DOWNES BURKE

TO

Question 1. There has been a lot of talk about reforming the USF contribution
assessment system. Much of this discussion has focused on moving toward a num16 The

model is sold by CostQuest, a company located in Cincinnati, Ohio.


issues arise between competing wireline carriers. Cost models must make assumptions about the sharing of facilities that are used in common.
17 Similar

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00149

Fmt 6601

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

146
bers-based system that would assess a per-line fee on all working telephone lines.
Do you believe that this could be implemented in a way that would not harm lowvolume and low-income telecommunications consumers?
Answer. Yes, a numbers-based contribution method could be implemented without
harm to low-income and low-volume consumers, but some versions of a numbersbased plan could cause harm.
Currently USF contributions are made as a percentage of interstate and international retail phone bills. Some carriers, such as wireless carriers and VoIP providers, are allowed to use a safe harbor calculation that presumes that a fixed percentage of their traffic is interstate and subject to the USF charge.
A customer who avoids all usage charges for interstate and international calls
(toll-free customer) typically pays the USF charge only on the Subscriber Line
Charge (SLC). Currently the national average residential SLC is $5.81 and the current USF rate is 11.7 percent. Therefore, an average toll-free customer currently
contributes about $0.68 per month to USF.
Low-Volume Customers
Shifting the contribution basis from revenues to numbers would have two effects.
It would reduce contributions from customers who have large interstate or
international service bills but few telephone numbers. Such customers likely include many business customers who have substantial expenditures for interstate services such as toll services and interstate special access (point-to-point)
lines. Some residential customers also have substantial interstate or international toll calling bills.
It would increase contributions from customers who have modest interstate
service bills or many telephone numbers. Most customers who make few telephone calls would pay larger contributions.
Some proposals for shifting to telephone numbers also would impose the unit USF
charge on connections or special point-to-point circuits. This question should not
be overlooked.
For example, a T1 or DS1telephone line can be thought of as a single connection or as 24 voice-grade channels. A T3 or DS3 line can be thought of
as one connection or as 672 voice-grade channels. A contribution mechanism that
imposes some form of charge on such special access circuits would likely reduce
charges on low-volume customers. A plan that requires greater contributions from
larger capacity special access circuits would further reduce any potential harm to
low-volume customers.
As noted above, the current USF charge for a toll-free wireline customer is probably about $0.70 per month. Estimates of a numbers-based contribution methodology were recently filed with the FCC in its intercarrier compensation docket.18
That filing estimated the effect of collecting Universal Service revenues through a
surcharge imposed on 614 million telephone numbers and special access connections.19 Based on that estimate, current USF costs 20 could be covered by a uniform
surcharge of almost exactly $1.00. 21 Therefore a telephone number-based contribution method would likely increase payments by toll-free wireline customers by about
$0.30 per month.
Prepaid wireless users often have small monthly bills as well. The current USF
charge for a minimally used prepaid wireless customer is about $0.45 per month.22
A plan that would increase contributions to $1.00 per month would therefore roughly double the burden of USF charges on such customers. This could also produce
significant harm to the carriers that offer prepaid wireless service, reducing the cost
advantage they currently enjoy as against post-paid wireless plans.
18 See Letter to Marlene Dortch, Secretary of FCC, January 30, 2007, in CC Docket No. 01
92, Missoula Plan Amendment to Incorporate a Federal Benchmark Mechanism, filed by Indiana
Utility Regulatory Commission, Maine Public Utilities Commission, Nebraska Public Service
Commission, Vermont Department of Public Service, Vermont Public Service Board, Wyoming
Public Service Commission.
19 Lifeline customers were excluded.
20 The second quarter USAC report shows high-cost programs costing, on an annualized basis,
$4.35 billion, and the entire USF program at $7.36 billion.
21 The intercarrier compensation filing estimated a charge of $0.95 producing $6.97 billion.
22 Prepaid wireless customers ordinarily have monthly bills of about $10. Under the FCCs
safe-harbor rule for wireless, 37.1% of such retail bills are considered interstate. At the current
USF rate of 11.7%, a prepaid wireless customer would pay $0.43 per month in USF charges.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00150

Fmt 6601

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

147
Low-Income
Many low-income wireline customers limit their toll calls, and they can also be
expected to avoid the more expensive bundles that include unlimited toll services.
Therefore, a wireline customer who is a low-volume customer is likely to be a lowincome customer as well. The same is true of wireless services. Prepaid wireless
services are substantially less expensive than post-paid subscriptions provided by
the major carriers. Prepaid wireless carriers assert that they serve predominantly
low-income customers.23 Therefore a change that increases the burden on low-volume customers is likely to increase the burden on low-income customers.
If a per-number contribution mechanism would indeed harm low-income customers, that harm might be mitigated or even offset by exempting Lifeline customers from paying the USF charge.24
Question 2. The concept of reverse auctions has been widely discussed as one solution to the problem of unchecked High-Cost Fund growth. How fast do you believe
a reverse auction program could be implemented? Why is it better than other approachessuch as study area caps or disaggregation? And, if implemented, what
sort of savings do you think reverse auctions would provide?
Answer. As I discussed above, current Universal Service policy takes a pro-competitive stance about the number of competitors who can receive support in a single
area. The Identical Support Rule (ISR) treats all ETCs as equal, but it has shown
itself incapable of restraining fund growth. I believe we should replace the ISR with
a plan that supports fewer carriers in high-cost areas. However, this will require
us to differentiate between the winners who will get support and the losers who will
not. This is a difficult and distasteful task because of the economic effects of Universal Service support. If we give Universal Service support to a single carrier, that
carrier will have an enormous competitive advantage. It may be sufficient to effectively block robust wireline competition in that local exchange market.
Auctions offer the conceptual possibility that we could select one or two networks
for support, but without the usual dilemmas. With auctions, the FCC might be able
to use a facially neutral process that rewards the most efficient competitor, but
without having to overtly pick an exclusive franchisee. In short, auctions offer a theoretical possibility that we could limit fund size in a way that minimally harms
competition and which lets the process, rather than the regulators pick the winners.
But I am not convinced that auctions can be implemented successfully as a full
replacement for the existing systems. We have received many proposals, but they
differ radically on fundamental points. These include:
Should auctions be held in all areas or only in competitive areas?
Should there be a separate wireless auction?
Should there be one winner or many?
What Carrier of Last Resort obligations should be imposed on winners?
In addition, I have several practical concerns about auctions:
An auction might simply fail. Bids might be higher than the current amount
of support, or there might be no bids at all.
Networks are interdependent, and awarding support through an auction process
disturbs that system. Bidders may be forced to assume that their competitors
wholesale services will always be available.
Two years ago the FCC was told by the Tenth Circuit Court of Appeals, for the
second time, that our current support mechanism for large carriers does not
provide demonstrably sufficient support. This is an important problem for rural
customers in many parts of the country who happen to be served by Bell companies. Auctions are not likely to address that issue fully.
Study area caps might be imposed to halt growth in the Fund size, but this should
be only a very short-term solution because it would perpetuate the inequities in the
23 For

example, TracFone Wireles, Inc. filed comments at the FCC in 2002 asserting:
As a prepaid wireless carrier, TracFone appeals to many low-income customers who are unable to pass a credit check or to afford security deposits required by other CMRS carriers, as
well as many wireline carriers. Approximately 11 percent of TracFones customers have annual
incomes of less than $15,000 and approximately 16 percent of TracFones customers have incomes under $25,000.
Comments of TracFone Wireless, Inc., filed April 22, 2002 in Federal-State Joint Board on
Universal Service, FCC Docket 9645.
24 This was a feature of the per-number mechanism mentioned above that was filed in the
FCCs intercarrier compensation docket by several state commissions.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00151

Fmt 6601

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

148
current system. To the extent that the current system provides too much support
to some small suburban carriers, and too little support to some large rural carriers,
a cap would likely prevent adjustments to those support levels. Networks are dynamic, and telecommunications policies are changing rapidly. Freezing the status
quo for Universal Service support would ignore the dynamic nature of this network
and make the systems current imbalances even worse over time.
Disaggregation is an interesting idea, and it may be desirable, but it is not a plan
for fundamental reform. First, most exchanges in the U.S. are already
disaggregated. The FCC simply mandated that support be disaggregated for areas
served by large, so-called nonrural companies. In these areas, changing line counts
every quarter increase or decrease the support received by ILECs and CETCs alike.
Therefore, disaggregation, even if it were mandated, would likely affect a small
number of customers.
Second, disaggregation does not always save money. While the carriers have
shown that it would reduce their support in particular cases, this has not been
shown to be generally true.
Third, disaggregation maintains the Identical Service Rule (ISR) but applies it
in a more geographically precise way. I explained above my concerns about the ISR.
In brief, it relies upon the wireline carriers exchange boundaries and the wireline
carriers exchange costs in order to calculate support for a carrier that often does
not have exchange boundaries and that has different costs. Making such a rule more
geographically sensitive might be an improvement, but it ignores more fundamental
issues.
Question 3. Can reverse auctions be implemented in a manner that is truly competitively and technologically neutral? Wouldnt such a plan inevitably mandate
technology-based winners and losers?
Answer. I am not sure that it is possible for any Universal Service mechanism
that distributes a finite support amount to be truly competitively and technologically neutral. Inevitably, Universal Service support presents a dilemma. We
cannot afford to pay support to all carriers, nor would that accomplish anything useful. Nevertheless, we consider it repugnant to provide support to only one or two
carriers, thinking that is inconsistent with neutrality.
The current Fund growth among CETCs illustrates, in my view, why we cannot
afford a solution that is truly competitively and technologically neutral. At most, I
would suggest that we should seek competitive neutrality within the constraints imposed by other more basic principles.
It is difficult to design an auction that has no technological bias. Wireline and
wireless services have different characteristics and deployment levels, and any such
difference could be disqualifying. For example, cell site backup batteries generally
cannot sustain operations for 12 hours without a recharge. Suppose an auction process required bidders to provide fully functional service for 24 hours following a
power failure. That requirement would prevent most wireless carriers from offering
bids and thus would not be considered as competitively neutral.
Even when an auction is facially neutral, the award of support to a single bidder
is itself likely to create a preferential result. Universal service support can provide
a significant competitive advantage to the carriers that receive it.
RESPONSE

TO

WRITTEN QUESTIONS SUBMITTED BY HON. DANIEL K. INOUYE


HON. BILLY JACK GREGG

TO

Question 1. In 1997, the FCC adopted the principle that its Universal Service policies should be competitively neutral. In explaining this principle, the FCC concluded that competitive neutrality means that Universal Service support mechanisms and rules neither unfairly advantage nor disadvantage one provider over another, and neither unfairly favor nor disfavor one technology over another. But it
seems that people have different views as to how that principle should be applied,
particularly when it comes to providing support for different kinds of communications platforms. As members of the Joint Board, do you believe that this remains
a valuable principle, and how should it be applied to competition both among and
between communications platforms?
Answer. Competitive neutrality remains a valuable principle of universal service.
All eligible telecommunications carriers (ETCs) should be eligible for USF support,
regardless of the technology used to provide service. However, many people confuse
the issue of competitive neutrality with the issue of equal per line support. The two
issues are not necessarily the same.
As originally conceived, all high-cost support for all carriers was to be based on
the forward-looking economic costs of serving each area, as determined by an econo-

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00152

Fmt 6601

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

149
metric model.1 As a result, the costs of serving a particular area would not be based
on any individual carriers costs, and the per line support available for serving a
customer in that area would be the same regardless of the technology used to serve
the customer. ETCs, using whatever technology they chose, would compete for this
support by competing for customers. Whichever ETC could provide high-quality
service in the most cost-effective manner would garner more support than other
ETCs. Under this approach new technologies could enter a market and compete on
equal footing with entrenched incumbents. In other words, legacy high-cost support
would not confer an unfair advantage to the incumbent carrier.
Unfortunately, this is not the way it has worked out. Support based on forwardlooking economic cost has never been extended to rural carriers. Instead, support
for rural carriers is still based on each rural carriers embedded (legacy) costs. In
addition, access replacement support for both rural and non-rural carriers is still
based on embedded costs.2 Under the equal support rule ETCs with very different
cost structures than an incumbent wireline carrier nevertheless receive the same
amount of per line support as the incumbent carrier. Ironically, as an incumbent
rural carrier loses lines, the amount of per line support goes up, increasing the per
line support for the competitive ETC as well.3
Even worse, ETCs rarely compete for support. Since the FCC adopted a policy in
1999 of supporting all lines of all ETCs in high-cost areas, wireless ETCs have been
able to receive support for providing supplementary lines in high-cost areas. In
other words, wireless ETCssometimes several wireless ETCs in the same high-cost
areaare able to draw new support while the incumbent does not lose any support.4
This has caused the amount of high-cost support to multiply in many high-cost
areas. For example, within the ATT/BellSouth service territory in Mississippi, ATT/
BellSouth draws $101.2 million in high-cost support, while sixteen competitive ETCs
draw $118.5 million.5
The Commission needs to address the equal support rule if support for all carriers
is not going to be based on forward-looking costs. As an interim step, the FCC
should require support in rural study areas to be based on each ETCs own costs,
capped at the costs of the incumbent carrier. While this action may trim some excess, it will do little to address the runaway growth of the Fund which is caused
by supporting multiple ETCs in high-cost areas. The high-cost fund can be placed
on a sustainable basis only by limiting the amount of support available for each
high-cost area, and (1) requiring the ETCs to compete for the limited amount of support, or (2) limiting the support to only one ETC within each area. The first option
would place the decision on which carrier wins the subsidy in the hands of the customer. The second option would require a regulatory authority to determine winners
and losers among ETCs. This could be accomplished through a reverse auction
mechanism.
Question 2. Section 254(c) of the Telecommunications Act of 1996 defines Universal Service as an evolving level of telecommunications services and also sets
forth criteria that the FCC considers when it decides which services qualify as supported services eligible for Universal Service support. At present, it is my understanding that the Universal Service Fund does not support broadband service. But
then, the question always arisesshould it? And if so, when? Do you think that
Universal Service should evolve to support broadband services, and if so, what
would trigger such a determination?
Question 2a. Given that the law defines Universal Service as an evolving level of
telecommunications services and given that the FCC has classified cable modem
and DSL services as information services, would the Congress need to change the
statute to make broadband eligible for support?
Answer. Section 254(c)(1) sets forth the criteria which the Joint Board must consider in determining whether to add services to the definition of universal service.
1 See, In re: Federal State Joint Board on Universal Service, CC Docket No. 9645, First Report & Order (May 7, 1997), 223; 273; 293295. Larger non-rural carriers were to be moved
to model-determined support first, followed later by smaller rural carriers.
2 Access replacement support is provided by the Interstate Access Support mechanism for
price cap carriers and Interstate Common Line Support mechanism for rate of return and average schedule carriers. These two mechanisms paid out $1.95 billion in support in 2006. The FCC
has never explained how it will transition away from these embedded cost mechanisms.
3 See, 47 C.F.R. 36.601 et seq.
4 In some instances wireless carriers were already providing wireless service without a subsidy
for many years. Upon becoming an ETC, the wireless carrier is showered with support dollars
simply for serving the customers that the carrier was already serving. In these cases, high-cost
support to wireless ETCs is truly found money.
5 Universal Service Administrative Company, Federal Universal Service Support Mechanisms
Fund Size Projections for the Second Quarter 2007 (Nov. 2, 2006), Appendix HC01.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00153

Fmt 6601

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

150
One of the most important of these criteria is that the service has through the operation of market choices by customers, been subscribed to by a substantial majority
of residential customers. 6 The last time the Joint Board considered adding
broadband to the list of supported services in 2002, only 7 percent of residential customers actually subscribed to broadband.7 According to the FCCs latest report on
advanced services, 43.6 percent of residential customers subscribed to broadband as
of June 30, 2006.8 I believe that residential subscribership to broadband has now
passed 50 percent, and it is time for the Joint Board to once again consider adding
broadband to the list of supported services. However, there may be statutory impediments to taking this action, as discussed in the answer to the next question.
Answer. Obviously, Section 254 is not a model of clear draftsmanship. Section
254(c)(1) states: Universal service is an evolving level of telecommunications services that the Commission shall establish periodically under this section, taking into
account advances in telecommunications and information technologies and services.
I believe the clear intent of Congress was to allow the definition of Universal Service to expand to include broadband when broadband services become a widespread
and essential part of the national telecommunications landscape. However, because
the FCC has defined cable, wireline and wireless broadband services as information
services with a telecommunications component, 9 broadband may not qualify for
inclusion in the definition of Universal Service since it is not a telecommunications
service as required by the current wording of Section 254(c)(1).
In Section 706 of the Act, the term advanced telecommunications capability is
defined as high-speed, switched, broadband telecommunications capability that enables users to originate and receive high-quality voice, data, graphics, and video
telecommunications using any technology. In the Second Report on Advanced Services the FCC defined advanced telecommunications capability and advanced services as services providing transmission speeds of more than 200 kilobits per second
in both directions (upstream and downstream).10 The FCCs usage of the term advanced services has generally been synonymous with the term broadband.
The linkage of Universal Service to advanced services is obvious in the wording
of the Act. Section 254(b)(2) states: Access to advanced telecommunications and information services should be provided in all regions of the Nation. Section 254(b)(3)
states: Customers in all regions of the Nation, including low-income consumers and
those in rural, insular, and high-cost areas, should have access to telecommunications and information services, including interexchange services and advanced
telecommunications and information services, that are reasonably comparable to
those services provided in urban areas and that are available at rates that are reasonably comparable to rates charged for similar services in urban areas. Section
254(b)(6) states: Elementary and secondary schools and classrooms, healthcare providers, and libraries should have access to advanced telecommunications services as
described in subsection (h) of this subsection. 11
The above-quoted portions of Section 254 appear to give the FCC sufficient authority to support advanced services. In fact, advanced services are currently supported under both the Schools & Libraries Fund and the Rural Health Care Fund
of the USF, even though advanced services or broadband are not currently included
6 Section

254(c)(1)(B) of the Act.


the Matter of Federal-State Joint Board on Universal Service, CC Docket No. 9645, Recommended Decision (July 10, 2002), at 13.
8 High Speed Services for Internet Access as of June 30, 2006, FCC Industry Analysis & Technology Division, Wireline Competition Bureau (Jan. 2007). As shown on Table 3, 50.2 million
residential customers subscribed to high-speed broadband services as of June 30, 2006. This represents 43.6 percent of the 115 million households in the United States.
9 See for example, In the Matter of Appropriate Framework for Broadband Access to the Internet over Wireline Facilities, CC Docket No. 0233, Report & Order & Notice of Proposed Rulemaking (Sept. 23, 2005), at 12; 105: The record demonstrates that end-users of wireline
broadband Internet access service receive and pay for a single, functionally integrated service,
not two distinct services. This conclusion also is consistent with certain past Commission pronouncements that the categories of information service and telecommunications service are
mutually exclusive. . . . We conclude now, based on the record before us, that wireline
broadband Internet access service is, as discussed above, a functionally integrated, finished
product, rather than both an information service and a telecommunications service.
10 In the Matter of Inquiry concerning the Deployment of Advanced Telecommunications Capability to All Americans, CC Docket No. 98146, Second Report (Aug. 21, 2000), at 11.
11 Section 254(h)(1)(B) requires that telecommunications services included in the definition of
Universal Service be provided to schools and libraries at rates less than the amounts charged
for similar services to other parties, while Section 254(h)(2) requires the Commission to establish rules to enhance, to the extent technically feasible and economically reasonable, access to
advanced telecommunications and information services for all public and nonprofit elementary
and secondary school classrooms, healthcare providers, and libraries . . ..
7 In

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00154

Fmt 6601

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

151
in the definition of universal service. While an argument could be made that
broadband can be supported under the current language of Section 254, to remove
all doubt it may be necessary to amend Section 254(c)(1) as follows: Universal service is an evolving level of telecommunications and information services that the
Commission shall establish periodically under this section, taking into account advances in telecommunications and information technologies and services. (New language in italics.)
Question 3. Currently the wireless eligible telecommunications carriers (ETCs) receive Universal Service support on a per customer basis based on the per line
costs of the wireline carrier in the same geographic area. This is sometimes called
the identical support rule and ensures that different communications platform providers receive the same amount of per line support. One criticism of the so-called
identical support rule for Universal Service is that it results in overly generous
support to wireless carriers because levels of support are not based on the per line
cost of providing wireless services. As a result, I have two questions
First, do you believe that Universal Service should support both wireline and
wireless services in rural America?
Answer. Please see my response to the first question above. Currently, both
wireline and wireless services are supported. During 2006 wireless ETCs received
$1 billion in high-cost support. As discussed above, the problem with the current
system is that it supports all lines of all ETCs in high-cost area without any restriction on the total amount of support and without requiring ETCs to compete for the
support. The result has been runaway growth of the high-cost fund, from $1.7 billion
in 1999 to $4.1 billion in 2006. Under either solution to this unsustainable growth
competition among ETCs or limitation of support to a single ETCwireless ETCs
should be allowed to compete on the same basis as wireline incumbent carriers.
Another alternative would be to create a separate wireless infrastructure fund,
similar to the Schools & Libraries Fund, to promote wireless build-out in rural
areas. Under this approach, a set amount of support would be made available each
year specifically to subsidize construction of additional wireless tower sites in rural
areas where such construction is currently uneconomic. Over time, a wireless infrastructure fund would ensure that customers in rural areas would have access to the
same level of ubiquitous wireless service as is enjoyed by urban customers.
Question 3a. Second, would it be possible to construct a model for wireless carriers
that would calculate support based on costs of wireless carriers, and what effect
might that have on the size of the fund?
Answer. While it would certainly be possible to construct a separate model for
wireless carriers, it would have little impact on the overall size of the high-cost
fund. As stated above, the main issue driving the size of the Fund is not equal per
line support, or how that per line support is calculated; it is the fact that the current system supports all lines of all ETCs in high-cost areas. It makes no sense to
subsidize two, three or more providers in areas where costs are so high that not
even a single carrier can provide service without a subsidy. As discussed in answer
to the first question above, support should normally go to the carrier that can provide service in the most cost-effective manner in high-cost areas, irrespective of
whether that carrier is a wireline carrier or a wireless carrier. As discussed below,
one way to get direct information on different carriers costs is head-to-head competition through reverse auctions.

RESPONSE

TO

WRITTEN QUESTIONS SUBMITTED BY HON. BILL NELSON


HON. BILLY JACK GREGG

TO

Question 1. There has been a lot of talk about reforming the USF contribution
assessment system. Much of this discussion has focused on moving toward a numbers-based system that would assess a per-line fee on all working telephone lines.
Do you believe that this could be implemented in a way that would not harm lowvolume and low-income telecommunications consumers?
Answer. No. Any move from a USF contribution system based on usage to a contribution system based on access will inevitably shift cost responsibility from highvolume users of telecommunications services to low-volume users. An access-based
contribution system, whether it uses numbers or connections, assesses every point
of access the same, regardless of the amount of usage through that point of access.
As a result, a customer with $1,000 of monthly usage through a point of access
would pay the same USF contribution as a customer with $0 monthly usage. While
it may be possible to mitigate the impact on low-usage customers by placing more

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00155

Fmt 6601

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

152
USF revenue responsibility on high-capacity data lines, the cost shift to low-usage
customers cannot be eliminated.
I believe a better approach to increasing the contribution base would be to remove
the current statutory restriction on assessing all revenues. Section 254(b)(4) states:
All providers of telecommunications services should make an equitable and nondiscriminatory contribution to the preservation and advancement of universal service. However, Section 254(d) limits this obligation to Every telecommunications
carrier that provides interstate telecommunications services. . . . The Fifth Circuit
Court of Appeals ruled in 1999 that the statutory language in Section 254(d) limits
the FCC to assessing only interstate revenues as the basis for contributions to the
USF.12 In 2003 the Joint Board forwarded to Congress recommended language to
broaden the FCCs assessment authority.13 I continue to support that recommendation. The proposed statutory changes to Section 254(d) are as follows: Notwithstanding the provisions of Section 152(b) of this Title, Eevery telecommunications
carrier that provides interstate telecommunications services shall contribute, on an
equitable and nondiscriminatory basis, to the specific, predictable, and sufficient
mechanisms established by the Commission to preserve and advance universal service. (Additions underlined; deletions of existing language marked by strikethroughs.)
Question 2. The concept of reverse auctions has been widely discussed as one solution to the problem of unchecked High-Cost Fund growth. How fast do you believe
a reverse auction program could be implemented? Why is it better than other approachessuch as study area caps or disaggregation? And, if implemented, what
sort of savings do you think reverse auctions would provide?
Answer. A reverse auction program could be implemented rather quickly if it was
done in the manner suggested by Verizon.14 Verizon proposes that the high-cost
fund first be capped by study area, with a separate cap for wireline and wireless
ETCs. Reverse auctions would first be conducted in study areas with multiple wireless ETCs to determine which single wireless ETC would receive support. Reverse
auctions could later be conducted between the wireline ETC and wireless ETC in
each study area (or within subdivisions of the study area) to determine which single
ETC would receive support.
As discussed above, reverse auctions would be used in conjunction with study area
caps to maintain stability in the Fund while the auctions were phased in. Over time,
reverse auctions should reduce the overall size of the Fund as support for multiple
ETCs within the same study area is eliminated. One of the most attractive features
of auctions is that they allow market forces to be injected into the USF support system. Periodic reverse auctions will also capture changes in technologies and underlying costs over time.
Savings achieved by auctions in study areas with excessive support could be used
along with disaggregation to address the issue of the current maldistribution of support among states and among study areas. In other words, while there are highcost areas which currently receive too much support, such as the example of Mississippi cited above, there are other high-cost areas that receive no support. This
is usually due to the fact that these high-cost areas are served by non-rural instead
of rural incumbent carriers. Updated computer models could determine the forwardlooking economic cost to serve every area in the United States, irrespective of which
incumbent carrier serves that area. These disaggregated costs could then be used
as the basis for reserve prices in future auctions.15 Reserve prices could be adjusted to fit whatever total amount of support is available.
Question 3. Can reverse auctions be implemented in a manner that is truly competitively and technologically neutral? Wouldnt such a plan inevitably mandate
technology-based winners and losers?
Answer. By its very nature, a reverse auction will be competitive and will inevitably determine winners and losers of the explicit USF subsidy for serving high-cost
areas. If price is the only criteria considered in a reverse auction, then the auction
process will favor those technologies with the lowest cost structures. This would
tend to favor wireless carriers in most parts of the Nation. However, in any reverse
auction the determination of the criteria that all bidders must meet is critical.
Under Section 214(e) of the Act, an ETC must be able to deliver all of the supported
services throughout the designated area. This may be difficult to achieve for many
12 Texas

Office of Public Utility Counsel v. FCC, 183 F.3d 393, 448 (5th Cir. 1999).
from the Federal-State Joint Board on Universal Service to the Hon. Conrad Burns,
dated May 19, 2003.
14 See Verizon ex parte filing with the FCC, February 9, 2007.
15 A reserve price is a price above which bids will not be accepted. Stated another way, a
reserve price is the highest level of support a regulator is willing to pay in a particular area.
13 Letter

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00156

Fmt 6601

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

153
wireless carriers, especially those that have not built out their networks in rural
areas. This is why I suggested above that a separate rural wireless infrastructure
fund may be the most appropriate way to ensure the build out of wireless service
to all parts of our Nation.
RESPONSE

TO

WRITTEN QUESTION SUBMITTED BY HON. DANIEL K. INOUYE


DAVID CROTHERS

TO

Question. There is a proposal before the FCC to restrain the growth of the Universal Service Fund by using reverse auctions. Under this proposal, carriers would
bid for the right to provide service in a given service area, for a given time with
the entity making the lowest bid winning the right to receive support. While I appreciate the benefits of reverse auctions, I also worry about potential costs like
lower service quality in rural areas, and the potential for creating stranded costs
for auction losers that might harm access to capital.
What effect would the possibility of losing support have on the ability of carriers
to attract private investment from capital markets?
What would happen if a provider wins the auction by bidding too much, and
then responds later by raising prices or reducing service quality?
What effect would reverse auctions have on those providers that fail to win support and their ability to roll out new services in rural America?
Answer. Senator, the North Dakota Association of Telecommunications Cooperatives (NDATC), the National Telecommunications Cooperative Association (NTCA)
and the universe of rural communications providers that are members of these two
organizations certainly share your concerns with regard to the general concept of
reverse auctions for the primary reason that they most certainly would lead to
stranded investment, placing systems that were built with Universal Service dollars
and Federal Rural Utilities Service loans at great risk of failure. Indeed, last October, NTCA commissioned a review of the subject of using reverse auctions to distribute Universal Service support which concluded that moving in this direction
with regard to areas with existing infrastructure and ubiquitous service would be
a serious mistake. The paper prepared by Alaska Pacific University Professor Dale
E. Lehman demonstrates the difficulties and dangers and inherent issues in applying reverse auctions in areas with existing infrastructure. Based on Lehmans findings, correctly designing and implementing an effective reverse auction mechanism
for rural markets will prove tremendously challenging, if not impossible.
According to the comments, while reverse auctions may be an appealing theoretical concept, their practical application is fraught with uncertainty and risk. Additionally, a reverse auction would be time and labor intensive, prohibitively expensive, and technically burdensome. The cost of administering the reverse auctions,
preventing fraud, and monitoring the results would ultimately increase the size of
the Universal Service Fund and could outweigh any potential benefits gained from
the process according to the author.
With regard to your specific question of what effect the possibility of losing support via an auction process would have on a carriers ability to attract private investment, we believe it is undeniable that the impact would be dramatic. For the Nations smallest carriers the impact would be particularly devastating. Today, policy
modifications, or even the potential of such modifications is turning our industrys
cost recovery ability and stability on its head. Rural carriers have traditionally not
been of a size that they are able to attract the interest of capital markets either
nationally or locally. This is why Universal Service is so important to rural carriers.
It is a cost recovery source, but it is also a necessary revenue stream that is essential to their ability to secure financing from the three primary sources of capital that
are available to them, the Rural Utilities Service, CoBank, and the Rural Telephone
Finance Cooperative (RTFC).
Your question regarding what would happen if a provider wins the auction by bidding too much and then responds later by raising prices or reducing service quality
is entirely justified. This is a fear that we outlined in our comments to the Federal
Communications Commission on the subject of reverse auctions. Indeed, it would be
virtually impossible to prevent this sort of scenario from playing out under a reverse
auction system. And as your question alludes, the real loser would not be the provider, but the consumer. Again, we think questions like this raise such dramatic
possibilities as to invalidate the concept from being considered any further whatsoever.
Finally, on the matter of what would happen to providers that fail to win support
under a reverse auction system, the response is very simple. They would quickly fail

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00157

Fmt 6601

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

154
and their most rural consumers would be those most hurt by the companys demise
because any new carrier would be highly unlikely to extend support beyond the easiest to reach consumers that reside within the town or community itself. There are
reasons why the Universal Service policy has evolved in the manner it has to best
serve rural markets and that is because they simply cannot be squeezed into an economic theory and be expected to work. During the extensive debate leading to the
development of the Telecommunications Act of 1996, NTCA and its Rural Telephone
Coalition partners, OPASTCO and WTA prepared and widely circulated a report titled Rural Is Different. While its underlying message was so simple and so obvious,
it was amazing at how hard we had to work to convince policymakers of its truth.
Sadly it appears that only 10 years later, many of your colleagues have already forgotten the reality of this message.
RESPONSE

TO

WRITTEN QUESTIONS SUBMITTED


DAVID CROTHERS

BY

HON. BILL NELSON

TO

Question 1. Is a reverse auction process the best way to reduce overall Fund
growth? What do members of the panel think of other options, such as breaking up
(or disaggregating) study areas to target funds to areas that are truly High-Cost?
Answer. Absolutely not! And we have been stating this fact ad-nausea for the past
decade. Frankly Senator, NDATC, NTCA, and their hundreds of members cannot
comprehend the reluctance of policymakers, either here on Capitol Hill, or at the
Federal Communications Commission (FCC) to look to and apply the most obvious
and simple remedies that would easily control the growth of the Universal Service
Fund. Repeatedly we have suggested four ideas to accomplish this objective:
1. Apply a meaningful public interest test when considering future eligible telecommunications carrier (ETC) designations;
2. Eliminate the identical support rule that today provides support to competitors based on an incumbents costs;
3. Provide alternative cost-based support to rural wireless ETCs; and,
4. Expand the base of USF contributors to include all broadband service providers.
Implementing this simple four pronged plan would immediately stem the flow of
Universal Service dollars and restore the confidence of all Americans in this timetested cornerstone of our national communications policy.
Question 2. If we move to a reverse auction process, isnt there a possibility that
some providers may bid so low that they end up financially unable to provide service? Furthermore, if an auction winner went bankrupt, how can we be sure that
households in that area continue to receive service?
Answer. Sir, these are exactly the sorts of questions we posed in our filings to
the FCC and in the Lehman paper referenced in our response to Senator Inouyes
questions on this subject. That paper was formally filed with the hearing record of
this committee. At any rate, we think this is the sort of gamesmanship that could
easily take place under the reverse auction scenario. At the very time when so many
of your colleagues appear to be concerned about waste, fraud, and abuse with this
or any other Federal oriented program, we think it would be unjustified to move
in the direction of a concept such as this that sounds interesting in theory but falls
apart immediately when looked at as a serious option. And of course you are right
consumers would be the ultimate losers in this scenario. Certainly, if the incumbent
carrier had lost the auction to a low bidder, our viewpoint would be that the incumbent no longer has a responsibility of carrier of last resort obligations because they
would not have the financial resources to make such a commitment. And, the entity
that submitted the unrealistically low bid would be unable to fulfill the commitment
as well, so there really would not be a good option for consumers in such a situation.
That is why we believe reverse actions have no place in this discussion.
RESPONSE

TO

WRITTEN QUESTIONS SUBMITTED BY HON. DANIEL K. INOUYE


BRIAN K. STAIHR, PH.D.

TO

Question 1. There is a proposal before the FCC to restrain the growth of the Universal Service Fund by using reverse auctions. Under this proposal, carriers would
bid for the right to provide service in a given service area, for a given time with
the entity making the lowest bid winning the right to receive support. While I appreciate the benefits of reverse auctions, I also worry about potential costs like
lower service quality in rural areas, and the potential for creating stranded costs

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00158

Fmt 6601

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

155
for auction losers that might harm access to capital. What effect would the possibility of losing support have on the ability of carriers to attract private investment
from capital markets?
Answer. Telecom networks are highly capital-intensive to operate, particularly in
rural areas, and investors seek a commensurate level of security to offset the risk
associated with the costs of network investments over time. In markets that are otherwise uneconomical to serve, Universal Service support is an important part of that
cost recovery. Less certainty over the continued receipt of Universal Service support
will translate into a higher cost of capital for telecom operators. In the case of reverse auctions, that uncertainty would have to be addressed by rules that ensure
support will be specific, predictable and sufficient.
Question 1a. What would happen if a provider wins the auction by bidding too
much, and then responds later by raising prices or reducing service quality?
Answer. Ensuring appropriate network investment, service quality and comparability of pricing is a challenge facing any reform of Universal Service programs.
In the case of reverse auctions, we assume participants would be bidding to (at
least) meet minimum requirements in each of these categories, as well as living up
to the carrier-of-last-resort (COLR) requirements currently imposed on the incumbent carrier. We also assume there would have to be a failsafe mechanism to prevent a winning bidder from defaulting on those requirements.
Question 1b. What effect would reverse auctions have on those providers that fail
to win support and their ability to roll out new services in rural America?
Answer. For the rural areas that are truly uneconomic to serve, the withdrawal
of support would lead to a substantial elimination of new investment and likely discontinuance of services by those providers, because the ability to recover costs and
earn a reasonable economic return on investment would no longer be there. While
some rural town centers might continue to receive service from a carrier that failed
to win support, those in the outlying areas would have to rely on the provider that
did win support for service, and it would only be fair to transfer the carrier of last
resort (COLR) requirement to the new provider (which would ultimately require cooperation with, or preemption of state authorities). For this reason, it is very important to target support to the geographic areas that need it the most.
Question 2. Dr. Staihr, I was interested in your testimony arguing that support
should be provided on a more granular basis. Is this type of granular analysis administratively feasible and what steps would the FCC need to take to institute such
a model?
Answer. Especially in recent years, with advancement in computing technologies,
mapping software and the online availability of free mapping information, a granular analysis has become administratively feasible. Embarq has demonstrated this to
its satisfaction by performing granular analysis on some of our serving areas for
proof-of-concept purposes, and it was affirmed by two other witnesses at the February 20, 2007 en banc hearing of the Federal-State Joint Board on Universal Service. In an April 12 filing to the Joint Board, Embarq outlined 5 steps that the FCC
could take to gather the information necessary for such a model:
1. Collect population density data from companies choosing to submit such data
for study purposes;
2. Validate the population density data using Census data and establish the
need for granular analysis;
3. Collect customer location data from the companies that qualify for granular
analysis;
4. Select a suitable model for estimating cost of service; and
5. Identify the high-cost areas at a granular level using the selected model and
submitted data.
These steps are explained in more detail, beginning on page 160 of our ex parte
presentation, which Ive attached, and we would be happy to discuss our proposal
in further detail at your convenience.
Question 2a. Also, if distributions were made on a more granular level, what effect
would it have on the overall size of the Fund and on the distribution of funds among
carriers?
Answer. Making distributions on a more granular level would create countervailing pressures on the size of the Universal Service Fund. On the one hand, an
appropriately targeted fund, by dispensing with statewide averaging for some carriers, would bring support to some very rural areas that currently receive no support at all, replacing unsustainable cross subsidies with explicit support.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00159

Fmt 6601

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

156
At the same time, considering town centers separately from outlying areas could
eliminate many of the most egregious arbitrage and windfall opportunities that are
causing the Fund to grow out of control today. Ultimately, the impact on the Fund
size would depend on the particular choices made in implementing more granular
targeting. Eliminating the windfall opportunities and supporting rural areas independent of statewide averaging would create a much more equitable distribution
among carriers and allow universal services distributions to be more closely aligned
with economic costs.
Question 2b. What steps would states and/or the FCC need to take to do this kind
of mapping?
Answer. Embarqs proposed steps are outlined in our ex parte, attached to this
document. We have proposed a cooperative system where carriers could share data,
but USAC could also gather publicly available data on population density and other
factors affecting network costs.
RESPONSE

TO

WRITTEN QUESTIONS SUBMITTED BY HON. BILL NELSON


BRIAN K. STAIHR, PH.D.

TO

Question 1. Is a reverse auction process the best way to reduce overall Fund
growth? What do members of the panel think of other options, such as breaking up
(or disaggregating) study areas to target funds to areas that are truly High-Cost?
Answer. Ive attached Embarqs April 12 ex parte detailing our proposal for more
granular targeting of support, which has the benefit of eliminating some of the
windfall and arbitrage opportunities that are causing the Fund to grow, and ensuring that support flows to the most high-cost areas.
We believe such granular targeting is an important component of any attempt to
reform Universal Service, regardless of whether the FCC pursues reverse auctions,
provides explicit support for broadband, limits support to one carrier per geographic
area or addresses many of the other difficult issues at hand.
After the March 1 Commerce Committee hearing and the tremendous focus on
support for broadband, we believe that if Congress or the FCC made the decision
to explicitly support broadband, identifying and targeting to the most high-cost
areas would be an indispensable step to lay the foundations for such a move.
Question 2. If we move to a reverse auction process, isnt there a possibility that
some providers may bid so low that they end up financially unable to provide service? Furthermore, if an auction winner went bankrupt, how can we be sure that
households in that area continue to receive service?
Answer. Both good points. We assume any reverse auction system would have to
include a qualification system to ensure that those who bid for support are capable
of meeting the carrier-of-last-resort (COLR) requirements for that area, and are financially stable enough to minimize the risk of bankruptcy. In any event, any kind
of USF reform would need a fail-safe mechanism to ensure that consumers and local
businesses do not lose service.
EMBARQTM CORPORATION
April 12, 2007
Commissioner DEBORAH TAYLOR TATE,
Federal Chair, Federal-State Joint Board on Universal Service
Federal Communications Commission
Washington, DC.
Commissioner RAY BAUM,
State Chair, Federal-State Joint Board on Universal Service
Oregon Public Utilities Commission
Salem, OR.
Ex Parte Presentation
RE: HIGH-COST UNIVERSAL SERVICE SUPPORT, WC DOCKET 05337; FEDERAL-STATE
JOINT BOARD ON UNIVERSAL SERVICE, CC DOCKET 9645.
Dear Commissioner Tate and Commissioner Baum:
Embarq strongly supports the substantial and continuing efforts of the FederalState Joint Board on Universal Service (the Joint Board) to reform the Federal Universal Service Fund (USF) so it may better advance the Universal Service goals set
forth in the Telecommunications Act of 1996. Telecommunications markets have
changed substantially in the decade since the current Federal USF was created and,
accordingly, substantial reform is necessary to accomplish those goals. To this end,
the Joint Board can best ensure that USF reform fulfills the statutory goals for Universal Service by recommending that the Federal Communications Commission:

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00160

Fmt 6601

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

157
1. Focus on correcting the structural problems caused by the multiplicity of support recipients and the misallocation of support;
2. Stabilize the current system of Universal Service support;
3. Limit the duration of a freeze or cap so as to make it temporary;
4. Initiate a study to identify the highest-cost areas at a granular level; and
5. Follow a clear and achievable process to complete the study, and then provide
support dollars to the areas identified by the study.
If the Joint Board recommends these steps, and the Commission adopts them,
Federal USF will become the specific, predictable and sufficient 1 program called
for in the statute. Federal USF finally will provide explicit 2 support to those highcost areas where it is truly uneconomic to provide servicethat is to say where the
marketplace conditions would not provide sufficient incentives for any carrier to
offer service. This, in turn, will ensure that quality services [are] be available at
just, reasonable, and affordable rates 3 that are that are reasonably comparable
in rural and urban areas.4
I. The Joint Board Should Recommend Correction of the Structural
Problems Caused by the Multiplicity of Support Recipients and the
Misallocation of Support
There is widespread recognition that the current USF suffers from significant
structural problems. In particular, the current USF does not satisfy important statutory criteria set forth in Section 254 of the Communications Act. It does not provide specific, predictable, and sufficient support in all (or even most) high-cost areas.
The Federal USF does not adequately preserve and advance universal service, and
it continues to rely on implicit rather than explicit support through extensive use
of cost averaging in the face of competition that renders such an approach
unspecific, unpredictable, and insufficient. Finally, the Federal USF does not ensure
access to supported services at rates that are affordable, reasonable, and comparable
to rates in urban areas.4
At the outset, Embarq emphasizes that USF reform need not impact many carriers, such as many small and mid-sized, rural incumbent local exchange carriers
(ILECs), that are unaffected by the structural problems identified below. Indeed,
these carriers would retain all of their current options under Embarqs proposals in
this document, which would not necessarily alter USF treatment for those carriers.
In particular, the study to more accurately identify high-cost areas to support that
which Embarq proposes herein would be voluntary and any new support provided
to previously-overlooked areas would come directly from correcting the structural
problem of duplicative support. The study and related granular targeting of support
would not, therefore, necessarily disturb USF treatment of currently-supported
ILECs.
The record in this docket contains hundreds of filings, a great many of them detailing the problems and the urgent need for reform, and the Joint Board itself identified the problems and the need for reform at its last en banc hearing.5 This evidence and analysis leads inexorably to the conclusion that the Joint Board should
recommend, and the Commission should promptly reform two critically important
structural flaws in Federal USF.
1. Duplicative support is being awarded to multiple competitive eligible telecommunications carriers (CETCs) operating in a single market area. This policy
has been the primary source of excessive growth in USF support, as noted by
Chairman Martin and others.6 The multiplicity of support and excessive USF
growth harms consumers everywhere by increasing both the cost to provide
service and the aggregate demand for USF contributions.
2. At the same time, however, many of the highest-cost areasmany designated
as rural and many others designated as non-ruraldo not receive sufficient high-cost support. This was confirmed by the United States Court of Appeals for the Tenth Circuit 7 for the non-rural fund, and it is equally true for
1 47

U.S.C. 254(b)(5).
U.S.C. 254(e).
3 47 U.S.C. 254(b)(1).
4 47 U.S.C. 254(b)(3).
5 Federal-State Joint Board on Universal Service En Banc Meeting February 20, 2007.
6 Opening remarks of Chairman Kevin Martin, Federal-State Joint Board on Universal Service
En Banc Meeting February 20, 2007.
7 Qwest v. FCC, 398 F.3d 1222, 1234 (10th Cir. 2005).
2 47

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00161

Fmt 6601

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

158
many carriers that receive support under the rural fund due to the current
practice of using averages (on a statewide or study area basis) to determine the
need for support.8 This failure to direct specific, predictable, and sufficient support to all areas that are truly uneconomic to serve harms consumers by inhibiting network investment in high-cost areas and perpetuating implicit subsidies
in lower-cost areas.
The Joint Board can best accomplish its objectives by issuing a Recommended Decision that focuses on steps to eliminate these structural flaws. In particular,
Embarq agrees with Windstream that the Joint Board should recommend forwardlooking and rational Universal Service reforms that target adequate explicit support
to high-cost areas. To do otherwise, would perpetuate the inequities and
insufficiencies in the current mechanism to the detriment of rural consumers and
the Nation. 9 Moreover, by fixing this structural flaw, the Commission can finally
comply with statutory mandates and the remand in Qwest v. FCC.
II. The Joint Board Should Recommend Stabilizing the Current System of
Universal Service Support
The first step to fixing the USF structural flaws is to prevent further harm, and
to do so sooner rather than later. The current growth in support, particularly increases that fund competition in areas where it is uneconomic for a single provider
to offer service, harm consumers and investment. AT&T and Verizon 10 have each
recently filed plans addressing this issue. Both of these plans propose that USF reform occur in two stages: (1) imposing a temporary freeze or cap on USF distributions to stabilize the system and permit the Commission to address current concerns
regarding fund size, fund growth, and magnitude of contribution factor; and (2) restructuring the method by which USF support is distributed.
As the first phase of a two-phase plan, a temporary freeze or cap would accomplish the important goal of immediately eliminating any additional upward pressure
on the end-user USF assessment, which is currently up to 11.7 percent. For the past
4 years, the overwhelming majority of the growth in high-cost support has been
driven by growth in wireless receipts while wireline receipts having stayed constant
or declined.11 This has happened because wireline support has long been subject to
a cap. Therefore, the most direct and narrowly-tailored, and competitively-neutral
approach to the problem is to address wireless support during this interim period.
Given that the purpose and justification for a temporary freeze or cap is to support longterm reform, it is critical that the freeze or cap be accompanied by a study
to identify the truly high-cost areas in the United States. The Joint Board should
recommend, therefore, that the Commission conduct such a study during the course
of a temporary freeze or cap. The public interest is best served through informed
decisionmaking, which can only be helped through a study of the cost of providing
service. In fact, this information is vital to any reform the Commission may consider, as explained below. A temporary freeze or cap will help ensure that the study
results are relevant (the freeze or cap will help maintain the conditions that will
be revealed through study) and accurate (the freeze or cap will help minimize gaming).
All other things being equal, a temporary freeze would be preferable to a cap in
economic terms since it ensures that no individual recipient would be made any
worse off or any better off as a direct result of the freeze during the interim timeframe. Conversely, a cap on funds may allow for the possibility of individual winners and losers underneath the cap as relative support amounts continue to be adjusted. This would be undesirable from a policy perspective as it would make study
results less accurate and relevant to the Commissions objectives.
III. The Joint Board Should Recommend That Any Freeze or Cap Be
Temporary
There are, of course, some risks involved in implementing any type of freeze or
cap; one being the natural tendency to apply a temporary remedy and then act as
if the problem has been solved. The Joint Board must emphasize, therefore, that
8 E.g., letter from Eric N. Einhorn, Windstream, to Deborah Taylor Tate, FCC and Ray Baum,
Or. Pub. Serv. Cmmn, WC Docket No. 05337 filed April 2, 2007 (Windstream Ex Parte).
9 Windstream Ex Parte, at 3.
10 See Letter from Robert W. Quinn Jr., AT&T, to Deborah Taylor Tate, FCC and Ray Baum,
Or. Pub. Serv. Commn, WC Docket No. 05337 filed March 22, 2007 (AT&T Ex Parte). See
also letter from Kathleen Grillo, Verizon, to Deborah Taylor Tate, FCC and Ray Baum, Or. Pub.
Serv. Commn, WC Docket No. 05337 filed February 9, 2007 (Verizon Ex Parte).
11 See Letter from Jamie M. Mike Tan, AT&T, to Marlene Dortch, FCC in WC Docket No.
05337 filed April 2, 2007.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00162

Fmt 6601

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

159
any temporary freeze or cap is a means to an end, rather than an end in and of
itself. Accordingly, Embarq agrees with AT&T when it proposes strict time limitationsa maximum of 2 yearson the duration of any freeze or cap.12 A freeze or
cap of any longer duration would only perpetuate the implicit subsidies that plague
the current USF.13
The Commission has the authority to impose a temporary freeze or cap, particularly in a case like this where the Commission requires market stability while it
studies where and how to best allocate USF support to the high-cost areas that most
need it. The implementation of a temporary freeze or cap on USF support is logical
because it is imperative that the Joint Board and the Commission address the underlying structural problems that are inherent in the current USF system. A temporary freeze or cap will provide the Joint Board and Commission with the necessary stability and time needed to accomplish this structural reform in a manner
that ensures the ongoing sufficiency, specificity and predictability of the Federal
mechanism.
The Commission enjoys considerable discretion to adopt interim rules while it undertakes long-term changes to its regulations. This is particularly so where the interim rules merely maintain the status quo so that the objectives of a pending rulemaking proceeding will not be frustrated. 14 In the case of USF reform, a temporary
freeze or cap is particularly appropriate given the rapid increases in overall support
and the substantial changes in support levels for individual carriers, including substantial decreases in support for some carriers. As the United States Court of Appeals for the District of Columbia Circuit has explained, [a]voidance of [such] market disruption pending broader reforms is, of course, a standard and accepted justification for a temporary rule. 15
IV. The Joint Board Should Recommend a Study To Identify the HighestCost Areas at a Granular Level
During the course of a temporary freeze or cap, the Commission will be in a position to undertake a detailed study that will identify the best means for addressing
the structural problems identified above. Windstream is correct when it observes
that the public interest will not be served by perpetuating the current system,
which is rife with inequities and logical failings. Therefore, the Joint Board should
recommend solutions for both structural problems discussed aboveduplicative support in some areas and inadequate support in others. One approach to solving both
problems would be to direct support freed up by fixing the duplicative support problem toward fixing the inadequate support problem.
As described at length during the Joint Boards recent en banc on universal service, the ability to accurately identify high-cost areas at a very granular level has
reached a level of precision that was unimaginable only a few years ago.16 Through
a combination of advances in modeling, better data, and ever-increasing computing
power the Commission has at its disposal a set of tools capable of producing a study
to ensure that all high-cost areas that truly require explicit support are adequately
supported. This is in stark contrast to the data and modeling capability that was
available nearly 10 years ago, when the Commission and Joint Board first considered using a study to determine USF needs.
A study would support, and would be a necessary precondition to implementing
a proposal like, AT&Ts. AT&T states as much in its ex parte presentation where
it wrote that, in order to ensure sufficiency of support to all high-cost areasincluding areas that do not currently receive high-cost support due to averagingit is necessary to determine the need for support at a more granular level (. . . in narrower
geographic areas, such as wire centers or Census Block Groups).17 The Joint Board
should recommend this be done by undertaking a comprehensive study that more
accurately identifies high-cost areas at a wire center or sub-wire center level.
A study would also facilitate and accelerate the implementation of any recommendation along the lines of a proposal like Verizons. Should the Commission
ultimately choose auctions as the best mechanism for addressing the problem of duplicative support awarded to multiple CETCs in a single area, it is important that
the Commission identify the areas that most need support. Conducting a study
12 AT&T

Ex Parte.
47 U.S.C. 254 (directing that implicit subsidies be made explicit).
Telecoms. Corp. v. FCC, 750 F.2d 135, 141 (D.C. Cir. 1984); see also CompTel v. FCC,
117 F.3d 1068 (8th Cir. 1997).
15 CompTel v. FCC, 309 F.3d 8, 14 (D.C. Cir. 2002) (citing MCI Telecoms. Corp., 750 F.2d at
141; ACS of Anchorage v. FCC, 290 F.3d 403, 410 (D.C. Cir. 2002)).
16 See presentation of Jim Stegeman, CostQuest Associates, Federal-State Joint Board on Universal Service En Banc Meeting, February 20, 2007.
17 AT&T Ex Parte, at 8.
13 See,

14 MCI

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00163

Fmt 6601

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

160
would help the Commission avoid many of the uncertainties and risks inherent in
using an untested approach such as reverse auctions to determine which areas
would be in need of support. The structural problems with the current USF make
it a poor guide for identifying the right areas to support. Moreover, it is important
to understand the costs of serving areas on a granular level in order to correctly
size the individual auction areas. Therefore, a granular understanding of which
areas are truly high-cost is essential to ensure that the areas to be bid on in any
auctions are those that best serve the public interest and fulfill the objectives of the
Communications Act.
In sum, a granular study would facilitate any long-term USF solution, and it
would do no harm. Moreover, a granular study identifying the truly high-cost areas
to serve will also produce the information needed by the Joint Board and Commission to evaluate future directions for the Federal USF mechanism and for USF policy in general. For example, the granular study would serve as an effective tool for
identifying areas where it is uneconomic for the market to deploy broadband. The
Commissions long-stated goal of advancing broadband deploymentwhether as a
supported service or notrequires a comprehensive understanding of the geographic
hurdles (density, distance, absence of critical mass of consumers) and incremental
investment needs that currently providers face as they bring advanced services to
the most rural, high-cost areas.
V. The Commission Should Follow a Clear and Achievable Process To Study
High-Cost Areas, and Then Provide Support Dollars to the Areas
Identified by the Study
The actual process for conducting a study to identify high-cost areas in need of
USF support is clear and achievable. First, Embarq proposes that the Commission
should maintain the support rules for companies that choose not to submit data for
a study. Then, the Commission should follow a five-step process to study high-cost
areas and identify new areas that should receive support. Finally, the Commission
should use study results to direct adequate support to the newly-identified high-cost
areas.
A. The Commission Should Maintain the Support Rules for Companies That Choose
Not To Submit Data for a Study
Embarq proposes that ETCs have the option not to participate in the study. Such
ETCs would continue to receive support as they do today. They would, however, remain subject to any applicable reforms, such as auctions (which may only apply to
a subset of ETCs under some of the proposals before the Joint Board). It is also important to note that a solution to this structural problem concerns identifying which
areas should receive support and directing to those areas the support that is currently misallocated due to the first structural flaw discussed abovesupporting duplicative ETCs. The question of which carriers should receive support and how that
support is to be calculated will be resolved in these new areas using the same methodologies that are chosen for currently-supported areas. In particular, ILEC costs
are used to identify high-cost areas today, and the study would follow this approach.
B. The Commission Should Follow a Five-Step Process To Study High-Cost Areas
and Identify New Areas That Should Receive Support
The Joint Board should recommend that the Commission determine that cost of
service is directly related to population density, and that study-area averaging masks
wide variations in the true cost of service. In general, low density translates to highcost. Because all network technologies (even wireless) exhibit economies of scale and
economies of density, there is a strong inverse relationship between cost and customer density. This relationship can be used to begin the process of accurately identifying high-cost areas. Many study areas exhibit a large degree of variation in density, which translates to a large degree of variation in costs. The current system of
using study area averages masks this variation in costs within a single study area.
In particular, the assumption that costs can be averaged is no longer valid because
of competition in low-cost areas, which prevents companies from realizing greater
margins in those areas and using those returns to support below-cost service in
high-cost areas.
The actual process for completing such a study is relatively straightforward, and
the study can be completed within the two-year time-frame of the freeze or cap. The
Joint Board should, therefore, recommend that the Commission take the following
actions:
1. Collect population density data from companies choosing to submit such data
for study purposes;

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00164

Fmt 6601

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

161
2. Validate the population density data using Census data and establish the
need for granular analysis;
3. Collect customer location data from the companies that qualify for granular
analysis;
4. Select a suitable model for estimating cost of service; and
5. Identify the high-cost areas at a granular level using the selected model and
submitted data.
1. Collect population density data from companies choosing to submit such data
for study purposes. In the first phase, if a company believes that the use of study
area averaging masks its high-cost areas, and therefore its need for USF support,
such companies could choose to submit disaggregated density data (for example, by
wire center or at a sub-wire center level) to the Universal Service Administration
Company (USAC). The National Exchange Carrier Association (NECA) could also
submit data on behalf of pooling companies that choose to participate but which
may not feasibly be able to submit their own data. The purpose of this showing
would be to demonstrate that significant variation in the density of areas served by
the carrier causes the carrier to experience significant variation in costs.18
2. Validate the population density data using Census data and establish the need
for granular analysis. USAC would independently verify this data using publicly
available Census data to determine whether the data showed significant variation
in density. If so, the strong density/cost correlation would allow USAC to conclude
that this area exhibited significant variation in costs (regardless of how costs might
be calculated). The preliminary evaluation would serve as an initial bright-line test
that this companys need for USF support must be determined at a more granular
level.
3. Collect customer location data from the companies that qualify for granular
analysis. At that point, a company that had initially submitted density data and
passed the bright-line test would then have the option of providing additional data
to USAC regarding wire center boundaries (just as it now provides Form 477 data
at a Zip Code level). The company would also have the option of submitting customer location data to USAC. Location data could be actual geo-coded locations, billing addresses, or service addresses.19 This data would remain proprietary and would
be held by USAC. It would be combined with public data (such as CB boundaries,
road systems) to be used to calculate costs (and ultimately, support).
4. Select a suitable model for estimating cost of service. Because companies actual
cost records do not generally exist at granular levels, it will be necessary to use a
model to estimate the cost of providing service of companies that choose to submit
the above-referenced data. The Commission would direct USAC to identify a model
that would most accurately estimate costs and partner with the models developer
on an ongoing basis to ensure that the use of the model would achieve the goals
set forth by Congress for Universal Service support mechanisms. Models are currently available that are capable of producing cost estimates for the entire country
at an extremely precise level, such as a single census block (CB) as identified by
the U.S. Census Bureau. To attain the level of accuracy necessary, the model must
incorporateto the greatest extent possiblereal-world engineering practices and
real-world network characteristics (such as road systems), as well as geo-coded customer locations into its forward-looking costing methodology.
5. Identify the high-cost areas at a granular level using the selected model and
submitted data. To determine which areas are uneconomic to serve and therefore
require support, the company-provided data (combined with publicly available data)
would be input into the selected model. Costs would be calculated and then produced at a level below the study area level to maintain a reasonable degree of granularity. Results would initially be produced at the individual wire center level,
which would yield an independently-identified list of high-cost wire centers that are
18 Until and unless a rule change is implemented that wireless carriers would receive USF
support based on something other than ILEC costs there would be no need for wireless carriers
to submit data. If such a change is made, competitive ETCs could, at their own choosing, also
submit density data regarding their designated service areas (which in many cases mirror existing study areas.)
19 Since wireless recipients are already required to provide line counts to USAC at the wire
center level all wireless companies that are USF recipients already have the capability of providing their customers locations by wire center even though they themselves do not operate
a network based on the concept of wire centers, if such a rule change occurred as described in
the footnote above.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00165

Fmt 6601

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

162
currently masked by the averaging process.20 This would give the Commission an
accurate compilation of high-cost areasin some cases entire study areas, in some
cases individual wire centers (or possibly zones)all of which are truly uneconomic
to serve and therefore in need of explicit support.
C. The Commission Should Use Study Results To Direct Adequate Support to the
Newly-Identified High-Cost Areas
Upon completion of the study, the Commission would still need to determine how
to provide adequate support to high-cost areas that are not currently receiving it.
In particular, the Commission would likely want to consider how support could be
provided to these areas without significantly increasing the size of USF. In the short
term the Commission could implement a pilot program to begin providing some level
of support to the highest-cost wire centers that had been identified by the study;
wire centers where the need for explicit support has been masked by the use of
study area averages. Funding for this support could come, for example, from AT&Ts
proposal for a 25 percent reduction in wireless receipts from the IAS and ICLS
funds.21
In the longer term, the answer can be found through a solution to the first structural problem listed abovethat of duplicative support going to multiple ETCs in
a single geographic area. To the extent the Commission undertakes action to reduce
the number of recipients in an areaand thereby reduce the dollars flowing to those
redundant ETCsthe existing support dollars that are freed up can be distributed
to the newly identified high-cost areas using the cost of providing service and an
appropriate revenue benchmark.
VI. Conclusion
In conclusion, the Joint Board can best ensure that USF reform serves the public
interest and benefits consumers by recommending that the Federal Communications
Commission: (a) focus on correcting the structural problems caused by the multiplicity of support recipients and the misallocation of support; (b) stabilize the current system of Universal Service support; (c) limit the duration of a freeze or cap
so as to make it temporary; (d) initiate a study to identify the highest-cost areas
at a granular level; and (e) follow a clear and achievable process to complete the
study, and then provide support dollars to the areas identified by the study.
Through this process, the Commission will accomplish all of its goals; it will:
Eliminate redundant, duplicative support;
Control fund growth; and
Identify accurately and direct support to all high-cost areas, including those
that have been overlooked because of the Commissions study-area averaging
approach.
Respectfully submitted,
BRIAN K. STAIHR, PH.D.
DAVID C. BARTLETT
JEFFREY S. LANNING
cc: Members and Staff of the Federal-State Joint Board on Universal Service
RESPONSE

TO

WRITTEN QUESTIONS SUBMITTED BY HON. DANIEL K. INOUYE


RICHARD N. MASSEY

TO

General Comments On Recent Joint Board Recommendation


The Joint Board recently adopted a Recommended Decision proposing a cap on
funding to competitive ETCs, while continuing to ensure that ILECs receive every
dollar that is currently disbursed to them. The Joint Boards proposal would cut
20 A carrier could also request the calculation of an added level of granularity. In many cases
there is significant cost variation within a single wire center, as described in Embarqs many
filings in this docket. This variation can be masked by the wire centers average cost, just as
wire-center-level variation often is masked within a study area average. A carrier requesting
increased granularity could request that the models results (which would have been calculated
by that time) be disaggregated to a more granular level, such as zones within a wire center.
This would be a very simple procedure because the actual model processing operates even more
granularly. For a company that requested additional granularity, the CB level costs could be
aggregated up to (for example) an inner- and outer-zone per wire center, based on contiguous
CBs above-or-below a certain density. The result, in this case, would be an independently-identified list of high-cost zones whose cost characteristics are currently masked by the averaging
process.
21 AT&T Ex Parte at 1011.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00166

Fmt 6601

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

163
funding to wireless and other competitive providers of Universal Service by 50 percent or more, while having no impact at all on funding to ILECs. This unfair and
anti-competitive recommendation effectively would hinder Universal Service by
making it harder for rural consumers to access the type of services that a majority
of consumers wantaffordable, high-quality mobile universal service.
The statutory principle of competitive neutrality prohibits the discriminatory approach recommended by the Joint Board, which both Democratic and Republican
members of this Committee have directly opposed. The Joint Boards recommendation flies in the face of S. 101, introduced in 2007 by Sen. Stevens, which properly
would codify the existing requirement that [u]niversal service support mechanisms
and rules should be competitively neutrali.e., that such rules must neither unfairly advantage nor disadvantage one provider over another, and neither unfairly
favor nor disfavor one technology over another. The version of H.R. 5252 adopted
by this Committee in 2006 included an identical provision. The Joint Boards recommendation also defies the request of Senators Rockefeller, Pryor, Dorgan, Klobuchar, and Smith, not to adopt a purportedly interim cap, especially one imposed only on certain carriers, because it would limit[] rural consumers options
and would impede the development of competitively neutral and even-handed interim and long-term reform measures. Likewise, the Joint Board ignored Senators
Sununu, McCain, DeMint and Ensign, who urged the Board not to adopt a plan
that would cap only one select group of providers but not another as we believe such
a fix would unfairly skew the marketplace. These Senators urged the Board not
to use interim measures, such as a temporary cap, and not to pick winners and
losers or favor one technology over another.
Alltel recognizes the widespread interest in controlling the growth of the Universal Service Fund. But such controls can and must be accomplished without compromising the principle of competitive neutrality or interfering with consumers access to wireless and broadband services. As Commissioner Copps recognized in his
testimony before this Committee, Bringing high-speed broadband to every corner
of the country is the central infrastructure challenge we face. In dissenting from
the Joint Boards recommendation, Commissioner Copps also expressed serious concerns that such a cap will be misinterpreted as a solution, even though it does not
addressor pretend to addressthe fundamental, comprehensive reforms needed to
carry a viable and improved system of Universal Service forward in the twenty-first
century.
The Universal Service system must be reformed in a manner that advances that
core objective, not in a way that obstructs the deployment of competitive broadband
facilities and services. Significantly, FCC data confirm that wireless carriers are
rolling out broadband services to consumers much more rapidly than any other telecommunications industry sector. Consumers in rural and high-cost areas would be
the ultimate losers under proposals that would substantially reduce or eliminate the
support needed to stimulate the deployment of wireless broadband networks and
services.
Question 1. There is a proposal before the FCC to restrain the growth of the Universal Service Fund by using reverse auctions. Under this proposal, carriers would
bid for the right to provide service in a given service area, for a given time with
the entity making the lowest bid winning the right to receive support. While I appreciate the benefits of reverse auctions, I also worry about potential costs like
lower service quality in rural areas, and the potential for creating stranded costs
for auction losers that might harm access to capital.
What effect would the possibility of losing support have on the ability of carriers
to attract private investment from capital markets?
What would happen if a provider wins the auction by bidding too much, and
then responds later by raising prices or reducing service quality?
What effect would reverse auctions have on those providers that fail to win support and their ability to roll out new services in rural America?
Answer. Alltel shares many of these concerns. A winner takes all auctionin
which only one provider could receive support funds at the end of the auction
would eliminate support for many wireless and wireline carriers that currently provide Universal Service throughout their designated service areas. This would make
it difficult or impossible for these wireless and wireline carriers to continue investing resources to provide high-quality, ubiquitous service in these high-cost areas.
Alltel believes that the public interest would not be served either by reverse auctions or by other changes to the high-cost funding ruleswhether characterized as
interim or long-termin which arbitrary reductions in support are imposed on
certain carriers or categories of carriers. Universal Service reformwhether through

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00167

Fmt 6601

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

164
auctions or some other reform measuremust be based on rational, well-supported,
analysis and decisionmaking.
Alltel has proposed a modest $25 million pilot program using reverse auctions to
promote broadband deployment in unserved or underserved rural markets that
would be designed to supplement, rather than replace, existing support mechanisms.
Under Alltels pilot proposal, and under any other form of competitive bidding process, reverse auctions should not be used to select a single ETC to receive support
in any geographic area, but only to set the amount of high-cost support funding per
line for all ETCs in each area. Winner takes all auctions would improperly distort
competition by having the government pick winners or losers. Instead, auctions
should be used, if at all, only to determine an efficient level of support that is the
minimum necessary to ensure the desired level of service in each geographic area.
Once that level of support is established, all carriers that satisfy the ETC requirements should receive the same (or comparable) amount of support per line, regardless of which one submits the lowest bid. In other words, rather than trying to use
competitive bidding as a substitute for actual competition, an auction-based funding
system could complement the competitive markets incentives for carriers to efficiently invest in rural markets and to provide high-quality service to rural consumers.
Such an auction structure would avoid distorting the marketplace after the auction is concluded and ensure that consumers receive the benefits of both Universal
Service and competition. It also would address concerns about a single auction winner later undermining Universal Service by raising prices or reducing service quality. If multiple ETCs are receiving funds and providing the supported universal
services after the auction, then market competition would protect consumers. If one
ETC were to raise prices or reduce service quality, then consumers could opt to purchase service from an alternative carrier that also receives the needed support to
serve the area. Also, an auction in which multiple carriers continue to receive support would reduce the likelihood that any one auction participant would offer an unreasonably low bid, because each bidder, as an ETC, would be required to provide
all the required elements of Universal Service to all consumers throughout the area,
consistent with 214(e)(1) of the Act.
Alltel opposes proposals, such as the Verizon plan, to use auctions for competitive
carriers, while retaining existing support mechanisms for the ILECs. These anticompetitive plans would likely eliminate funding for the majority of wireless carriers, lead to substantial reductions in funding to remaining wireless Universal
Service providers, while maintaining existing funding for wireline ILECs in most
cases. This outcome would unreasonably discriminate against wireless companies, in
violation of the Act and well established law, and to the detriment of consumers and
intermodal, facilities-based competition. It also would thwart efficient investment in
rural areas. Instead, if any reverse auction plan were adopted, it must be structured
to have all ETCs in a given geographic area participate in a single auction, regardless of the technology they use and regardless whether they are incumbents or competitive entrants.
Question 2. Mr. Massey, would it be possible to construct a Universal Service support mechanism for wireless providers that would be based on the cost of providing
wireless services?
In your view, what would be wrong with such an approach?
What effect would tying wireless support to wireless costs have on the size of
the fund?
Answer. Cost models could be developed to estimate the costs of providing service
in rural areas using both wireless or wireline technologies, and Universal Service
mechanisms could be developed to set support levels based on those costs. Ideally,
however, the Universal Service support mechanism would provide funding to every
ETC based on the cost of the most efficient (least cost) technology available to serve
all customers in the geographic areawireless or wireline. This would right size
the Universal Service Fund by preventing excessive disbursements to some carriers
just because those carriers have received large amounts in the past, while also ensuring that sufficient funds are available to enable carriers to provide the supported
services in high-cost areas. It also would create incentives for all carriers to operate
as efficiently as possible and would avoid giving discriminatory advantages to one
group of carriers and disadvantages to others.
Most importantly, a competitively neutral Universal Service systembased on the
costs of the most efficient technology, rather than based on an individual carriers
past investment decisionswould avoid distorting competition and would protect
consumers rights to select their preferred services in a competitive marketplace. By
contrast, it would make no sense to provide greater funding to more inefficient car-

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00168

Fmt 6601

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

165
riers (those that incur greater costs to serve consumers in a given area) and less
funding to more efficient carriers. Such a non-competitively neutral system would
create perverse incentives for carriers to operate as inefficiently as possible. It also
would discriminate against efficient service providers by depriving them of revenues
that are available to carriers that operate in a more costly manner.1
The 1996 Act requires that all Universal Service funding be competitively neutral.
In order to ensure competitive neutrality, all funds must be portablei.e., available regardless whether a consumer decides to purchase service from an ILEC or
a competitive ETCas West Virginia consumer advocate Billy Jack Gregg explained
in his March 1, 2007 testimony (see page 21), and as the FCC and the courts have
affirmed many times. This means that neither a wireline company nor a wireless
company should receive different amounts of support funding for providing service
to a given customer in a particular geographic area.
It would be inappropriate to depart from competitive neutrality by retaining current funding levels (based primarily on embedded or historical costs) for ILECs and
disbursing a reduced level of support to wireless or other competitive ETCs. The answer does not lie in trying to develop a new, separate set of rules for funding competitive ETCs, while allowing the ILECs to continue to operate under a monopolyinspired form of regulation, e.g., guaranteed rate of return on embedded costs, regardless of efficiency and effectiveness in serving rural areas. And it would be impossible as a practical matter to set wireless carriers funding based on their own
embedded costs, using cost studies that parallels the approaches used by the rural
ILECsi.e., using factors such as nationwide average cost per loop, subscriber line
charge revenue and DEM weighting. The application of these monopoly-oriented,
ILEC-based standards to wireless carriers would be a contrived and convoluted process, and ultimately would make no sense.
RESPONSE

TO

WRITTEN QUESTIONS SUBMITTED


RICHARD N. MASSEY

BY

HON. BILL NELSON

TO

Question 1. Is a reverse auction process the best way to reduce overall Fund
growth? What do members of the panel think of other options, such as breaking up
(or disaggregating) study areas to target funds to areas that are truly High-Cost?
Answer. In the very short-term, the best way to limit overall Fund growth would
be to adopt a competitively neutral proposal such as that advanced by West Virginia
consumer advocate Billy Jack Gregg. As an interim measure, Mr. Gregg has proposed a single inflation-adjusted cap on the growth of the total high-cost support
disbursed to all categories of ETCs (including ILECs and competitors). Funding
would be distributed among all eligible wireline and wireless carriers in each area,
with proportional adjustments based on each ETCs share of customer lines. Mr.
Greggs proposal would prevent undue growth in the overall level of funding while
also spreading the impact of the Fund growth limitation proportionately among all
ETCs. Unlike a wireless-specific fund cap, Mr. Greggs proposal would avoid severe
reductions in total support or per-line support to any category of carriers, would
avoid distorting competition or favoring one technology over another in rural areas,
and would avoid imposing barriers to entry.
In the medium- to long-term, Alltel agrees with Senators Rockefeller, Pryor, Dorgan, Klobuchar, and Smith that the Board should seriously consider competitivelyneutral proposals, ensure accountability for how funds are used, and promote advanced services in rural regions through effective targeting of funds to high-cost
areas. In particular, regulators could control Fund size while also advancing Universal Service more effectively by targeting funds to the highest cost disaggregated
geographic areas, regardless of whether those areas were served by a small, midsized, or large ILEC in the past. The current system disburses much more funding
to smaller ILEC study areas (even where the supposedly small ILEC operating
companies are owned by large holding companies), and improperly requires mid-size
and larger ILECs to support the high-cost portions of their study areas using implicit subsidies from low cost to high-cost areas. This system also harms wireless
ETCs such as Alltel that focus on serving consumers in rural areas. Instead, highcost support funding should be targeted to consumers in outlying rural areas.
1 For similar reasons, the public interest requires the elimination of the so-called rate of return system, in which some components of the existing Universal Service system reimburse
ILECs for each dollar they spend. This system creates perverse incentives for these carriers to
operate as inefficiently as possible, and unfairly guarantees these carriers revenue streams
while imposing marketplace risks on competitive carriers. The FCC has stated repeatedly, ever
since 1997, that it intended to eliminate this obsolete system, and Alltel filed a petition asking
it to do so in 2003. But thus far the FCC has failed to deliver on this commitment.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00169

Fmt 6601

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

166
Alltel also believes Fund growth can be controlled by imposing more rigorous
oversight to ensure that Funds are actually being used in a manner that furthers
the goals of the Universal Service Fund. Competitive ETCs are currently required
to submit detailed annual reports regarding their plans for network construction
and service quality improvement, as well as information on the amounts of Universal Service support received and how such support was used to improve their
networks and benefit consumers. However, in most states ILECs are not subject to
comparably rigorous reporting standardsbut they should be, in order to ensure the
integrity of the program. In addition, the oversight and processing of ILEC funding
should be entrusted to a neutral administrator subject to strict FCC oversight (i.e.,
USAC), rather than the rural ILEC-controlled advocacy organization (NECA) that
controls this process today.
Question 2. If we move to a reverse auction process, isnt there a possibility that
some providers may bid so low that they end up financially unable to provide service? Furthermore, if an auction winner went bankrupt, how can we be sure that
households in that area continue to receive service?
Answer. Alltel shares many of these concerns, and we have addressed them in the
response to Chairman Inouyes questions for the record. In short, we oppose a winner takes all auction, in which the auction would select a single ETC to receive
support and other ETCs would receive no support funds. Such a system would have
government pick winners and losers and would deprive consumers in these high-cost
areas of access to service from a range of competing service providers. Instead, reverse auctions should be used, if at all, only to set the efficient level of high-cost
support funding per line for all ETCs in each area.
Such an auction structure would avoid distorting the marketplace after the auction is concluded and ensure that consumers receive the benefits of both Universal
Service and competition. It also would address concerns about a single auction winner later undermining Universal Service by raising prices or reducing service quality. If multiple ETCs are receiving funds and providing the supported universal
services after the auction, then market competition would protect consumers. If one
ETC were to raise prices or reduce service quality, then consumers could opt to purchase service from an alternative carrier that also receives the needed support to
serve the area. Also, an auction in which multiple carriers continue to receive support would reduce the likelihood that any one auction participant would offer an unreasonably low bid, because each bidder, as an ETC, would be required to provide
all the required elements of Universal Service to all consumers throughout the area,
consistent with 214(e)(1) of the Act.
Question 3. Alltel and Verizon have presented separate proposals for a reverse
auction process. Can you explain what specifically makes your proposal superior?
Also, how do your plans differ from the reverse auction proposal presented by CTIA?
Answer. Alltel opposes Verizons plan and other proposals to conduct multiple separate auctions for different technologies or classes of carriers. These anti-competitive
plans would likely eliminate funding for the majority of wireless carriers, lead to
substantial reductions in funding to remaining wireless Universal Service providers,
while maintaining existing funding for wireline ILECs in most cases. This outcome
would unreasonably discriminate against wireless companies, in violation of the Act
and well established law, and to the detriment of consumers and intermodal, facilities-based competition. It also would thwart efficient investment in rural areas. Instead, if any reverse auction plan were adopted, it must be structured to have all
ETCs in a given geographic area participate in a single auction, regardless of the
technology they use and regardless whether they are incumbents or competitive entrants.
By contrast to Verizon, Alltel has proposed a modest $25 million pilot program
using reverse auctions to promote broadband deployment in unserved or underserved rural markets that would be designed to supplement, rather than replace,
existing support mechanisms. Verizons plan would continue to focus funding on traditional plain old telephone service; Alltels reverse auction plan is more forwardlooking because it would target funding to encourage deployment of new broadband
networks and services.
Alltels and CTIAs auction plans are similar in most respects. Both Alltel and
CTIA support auctions in which multiple ETCs would be able to compete in the provision of supported services after the auction concludes, and both Alltel and CTIA
oppose winner takes all auctions, for the reasons discussed above. CTIA has indicated that, if needed to encourage low bidding, the auction winner (i.e., the lowest
bidder) could receive slightly more funding per line than other ETCs in the area
(winner takes more). Alltel believes that it would be preferable for all ETCs to receive the same amount of funding, but would not object to CTIAs approach as long

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00170

Fmt 6601

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

167
as there is only a small difference between the amounts disbursed to low bidders
and to other qualifying ETCs.
RESPONSE

TO

WRITTEN QUESTIONS SUBMITTED BY HON. DANIEL K. INOUYE


THOMAS J. TAUKE

TO

Question 1. There is a proposal before the FCC to restrain the growth of the Universal Service Fund by using reverse auctions. Under this proposal, carriers would
bid for the right to provide service in a given service area, for a given time with
the entity making the lowest bid winning the right to receive support. While I appreciate the benefits of reverse auctions, I also worry about potential costs like
lower service quality in rural areas, and the potential for creating stranded costs
for auction losers that might harm access to capital. What effect would the possibility of losing support have on the ability of carriers to attract private investment
from capital markets?
Answer. There are several different proposals before the FCC regarding the use
of competitive bidding or reverse auctions to distribute Universal Service support.
Under Verizons proposal, the only carriers that could lose USF support are those
in areas where the system is supporting more than one network. Those carriers that
demonstrate their efficiency by winning the auction will continue to receive support
in the amount of their bids. Thus, the auction process itself will help capital markets identify efficient carriers, which could well promote private investment.
Verizons proposal would not flash cut to auctions, but would phase them in over
time, and would provide sufficient transitions for carriers that are currently receiving support. Verizon has proposed that the FCC phase in separate and parallel auctions: one auction in areas with more than one wireless provider receiving USF
funds and one auction in areas with more than one wireline provider receiving USF
funds.
Auctions initially would be held only among wireless ETCs and only in areas
where there is more than one wireless ETC. In this first phase, auctions would not
affect funding for rural telephone companies. After the wireless auctions have been
completed, the FCC would hold auctions among wireline ETCs in areas where there
is at least one wireline ETC. Because there are relatively few wireline CETCs today,
this part of Verizons proposal would affect very few areas. After both sets of auctions are completed, Verizon suggests that the FCC could assess the results of the
auctions held so far, and determine next steps.
Question 1a. What would happen if a provider wins the auction by bidding too
much, and then responds later by raising prices or reducing service quality?
Answer. Consumer choice is the most effective check on prices, and that would
not change if auctions are used to identify the most efficient carriers in high-cost
areas. More than ever before, consumers of communications services have options
from both traditional service providers and new offerings by cable, Voice over IP,
and wireless providersand they are taking advantage of them. Many of these providers operate without any Universal Service support, which constrains the prices
all carriers can charge. There may be some areas where wireline providers do not
face competition; in those areas, Verizon does not propose to hold USF auctions, and
in any event the auction process would not impact existing price regulations.
Every purchasing government agency that uses contractors must be concerned
with quality of service. In this context, as in the government procurement context,
the auction process itself can ensure that a supported provider offers a minimum
level of service. In an USF auction, a document like an RFP (request for proposal)
or an RFQ (request for a quote), which are used in other types of government procurement, would be issued. That document would define the obligations of the winning bidder in an auction. The bidder would know these obligations in advance and,
by bidding, would agree to accept them. Once the auction was over, the winning bidder would also sign a contract that would outline these responsibilities and which
would help ensure that service quality benchmarks are satisfied.
Question 1b. What effect would reverse auctions have on those providers that fail
to win support and their ability to roll out new services in rural America?
Answer. Auctions do not prevent carriers, even those carriers that participate in
but do not win the auction, from providing service in an area. Again, many providersespecially new intermodal providersoperate without any Universal Service
support, and would presumably continue to do so even in areas where an auction
has been held. Verizon supports targeting USF support to where it is truly needed;
in areas where a provider is able to operate without support, the presumption
should be that we do not need USF in that area to ensure that consumers have affordable access. Moreover, auction results would not stand forever. Carriers that do

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00171

Fmt 6601

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

168
not win the auction will have opportunities to bid again for support in the same
areas and to nominate other areas for auction.
We also must remember that todays Universal Service system, which bases support on a carriers costs, does not create ideal incentives for carriers to innovate and
develop new services. In contrast, an auction process would reward carriers for introducing new services because those carriers would have a stronger business plan
and would be better positioned to win an auction. Competition in the marketplace
has served American consumers well, and Verizons proposal would bring those
same incentives to bear for the benefit of consumers in rural areas.
RESPONSE

TO

WRITTEN QUESTIONS SUBMITTED


THOMAS J. TAUKE

BY

HON. BILL NELSON

TO

Question 1. Is a reverse auction process the best way to reduce overall Fund
growth? What do members of the panel think of other options, such as breaking up
(or disaggregating) study areas to target funds to areas that are truly High-Cost?
Answer. Disaggregation of support from study areas to wire centers or the subwire center level is one potential solution, but it is a potential solution for a different problemhow to better target high-cost funds to areas where they are truly
needed.
We should keep in mind that targeting to smaller geographic areas is not a way
to control the Funds size. On the contrary, if the current funding mechanism were
to be modified to use smaller geographic areas to distribute support, the result could
be a much larger fund.
Verizon is supportive of efforts to target support to areas where the need is greatest. In fact, our proposal makes it possible to target the funding to smaller geographic areas without making the Fund bigger, because we also suggest a cap that
provides immediate control of Fund growth and an auction mechanism that determines just the right amount of support for each targeted area. Gaining control of
fund growth through a reasonable cap is a critical first step that will give us breathing room to implement fundamental, long-term reforms.
Question 2. If we move to a reverse auction process, isnt there a possibility that
some providers may bid so low that they end up financially unable to provide service? Furthermore, if an auction winner went bankrupt, how can we be sure that
households in that area continue to receive service?
Answer. In any government procurement process, the responsible entity must ensure that the winning bidder will perform as specified in the contract. Auctioning
USF obligations is no different.
In the USF context, this can be accomplished by qualifying prospective bidders
to ensure that they are technically and financially able to perform, posting of bonds,
and the enforcement of penalties for nonperformance. Another enforcement mechanism could be disqualification from future bidding if a company fails to perform.
Question 3. Alltel and Verizon have presented separate proposals for a reverse
auction process. Can you explain what specifically makes your proposal superior?
Also, how do your plans differ from the reverse auction proposal presented by CTIA?
Answer. Verizon has proposed the most effective and workable path to Universal
Service reform. We propose immediate action in the form of reasonable caps at current funding levels to address the most immediate crisis the Fund faces: its rapid
and unsustainable growth. We propose implementing competitive bidding quickly
but on limited basis (first in areas with multiple wireless ETCs), and where it can
provide the greatest benefit. We then propose to give the Joint Board and the FCC
the flexibility to assess the results of these steps and to decide whether to extend
the reforms more broadly.
In contrast, Alltel does not propose a solution that will properly stabilize and rationalize the fund. Alltel proposes only to cap the per-line amount of support in each
area. This capping proposal would be ineffective. It ignores the main source of
growth in the fund: support provided to the growing number of wireless handsets.
Moreover, Alltel only supports the use of auctions for a small fraction of the Fund
($25 million) and only for new broadband services. If it is necessary for government
to intervene in broadband deployment, there are better ways to target broadband
support than by including broadband in the definition of services supported by the
current Fund.
CTIA is supportive of auctions, but suggests that auctions should be designed so
that multiple providers continue to receive support with the auction winner receiving a higher amount of support (which CTIA calls winner takes more.) However,
an auction that has more than one winner and no losers would neither rationalize
nor stabilize the system, and would not contain the growth of the fund.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00172

Fmt 6601

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

169
A winner takes more approach does not provide the proper incentives for participants to submit bids that are no larger than necessary to provide supported services, and could lead to collusion and strategic behavior that would skew the auctions results. For example, if all the bidders knew that no bidder could truly lose
the auction, there would be strong incentives for all the bidders to collude and submit large bids so that all of the participants received higher levels of support.
RESPONSE

TO

WRITTEN QUESTION SUBMITTED BY HON. DANIEL K. INOUYE


W. TOM SIMMONS

TO

Question. There is a proposal before the FCC to restrain the growth of the Universal Service Fund by using reverse auctions. Under this proposal, carriers would
bid for the right to provide service in a given service area, for a given time with
the entity making the lowest bid winning the right to receive support. While I appreciate the benefits of reverse auctions, I also worry about potential costs like
lower service quality in rural areas, and the potential for creating stranded costs
for auction losers that might harm access to capital.
What effect would the possibility of losing support have on the ability of carriers
to attract private investment from capital markets?
What would happen if a provider wins the auction by bidding too much, and
then responds later by raising prices or reducing service quality?
What effect would reverse auctions have on those providers that fail to win support and their ability to roll out new services in rural America?
Answer. As stated in my written testimony, the continued growth in the size of
the Universal Service Fund is a matter of significant concern to the cable industry
for a simple reasonthese costs ultimately are borne by consumers. Based on the
anticipated growth of cable telephony services, and the corresponding growth in the
share of the program that will be funded by cable consumers, our industry supports
efforts to reduce the burden of Federal support programs by more efficiently distributing support.
The above questions suggest concern about the impact reverse auctions would
have on existing networks. As network-based companies, we appreciate that concern. In reforming the program for distribution of Federal Universal Service support, however, it is important to keep in mind that the program was created to benefit consumers, not carriers. The subsidization of networks through a government
fund is simply a means to that end in situations where market forces would not otherwise meet consumers needs. Where market forces can meet those needs, as is increasingly likely given the growth of cable voice services, government subsidization
is unnecessary and potentially counterproductive.
Reverse auctions are a mechanism by which government can take advantage of
market forces (i.e., the presence of multiple networks in areas previously served by
a single network) to distribute support more efficiently. If structured properly, they
offer an opportunity not only to reduce the size of the fund, but also to promote competition in high-cost areas by making support available on a more equitable basis.
The challenge is to reduce the burden on consumers and promote competition, without sacrificing the level of service provided in these areas today. We believe that
an auction program can do this by specifying minimum levels of service to be offered
and establishing obligations to be met by all bidders. This should include some sort
of carrier-of-last-resort obligation, which will ensure that the fundamental goal of
providing service to all consumers is met. Any facilities-based provider that commits
to meeting these requirements should be eligible to participate in the auction.
Implementing a reverse auction process for Universal Service should not result in
stranded costs. If the auction takes place in an area with multiple networks, all
those networks have an incentive to compete for customers (because they need the
revenue to cover their costs) regardless of whether they win or lose the auction.
Even if an ILEC loses customers, the investment generally is not stranded because
it can be used if the carrier wins the customer back, which it has every incentive
to do.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00173

Fmt 6601

Sfmt 6621

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

170
RESPONSE

TO

WRITTEN QUESTIONS SUBMITTED


W. TOM SIMMONS

BY

HON. BILL NELSON

TO

Question 1. Is a reverse auction process the best way to reduce overall Fund
growth? What do members of the panel think of other options, such as breaking up
(or disaggregating) study areas to target funds to areas that are truly High-Cost?
Answer. As stated in response to Chairman Inouyes question above, we believe
that reverse auctions, if structured properly, offer an opportunity not only to reduce
the size of the fund, but also to promote competition in high-cost areas by making
support available on a more equitable basis. NCTAs view is that reverse auctions
can be effective only if they cover relatively small service areas. Not only is this critical to ensuring that the bidding process is competitively and technologically neutral, it also has the effect of targeting more support to truly high-cost areas while
reducing support to those areas where market forces are most active. NCTA would
not oppose consideration of other methods of targeting support if they could be accomplished in a manner that reduces overall Fund size.
Question 2. If we move to a reverse auction process, isnt there a possibility that
some providers may bid so low that they end up financially unable to provide service? Furthermore, if an auction winner went bankrupt, how can we be sure that
households in that area continue to receive service?
Answer. As noted above, NCTA believes that a minimum set of binding service
obligations should be part of any auction program. In addition, in establishing the
ground rules for such a program, the FCC could establish procedures to ensure continuation of service and to address the consequences of a bankruptcy filing.

VerDate Nov 24 2008

14:09 Mar 21, 2012

Jkt 073387

PO 00000

Frm 00174

Fmt 6601

Sfmt 6611

S:\GPO\DOCS\73387.TXT

SCOM1

PsN: JACKIE

Vous aimerez peut-être aussi